From joe at whipple.cc Mon Mar 1 09:17:42 2004 From: joe at whipple.cc (Joe Whipple) Date: Sat Jan 8 01:34:32 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <200402292351.58732.rfunk@funknet.net> References: <200402292351.58732.rfunk@funknet.net> Message-ID: <40434606.4090902@whipple.cc> I was under the impression samba already did this... There is a process named nmbd running when I start samba that does what you mention: root 20023 0.0 0.3 5312 1904 ? S 09:16 0:00 /usr/sbin/smbd root 20025 0.0 0.3 4120 1656 ? S 09:16 0:00 /usr/sbin/nmbd Joe Rob Funk wrote: >At the last COLUG meeting I attended, there was a little bit of side >discussion about trying to figure out a way for a Linux machine to use >Windows host resolution (NMB). I think I mentioned that the ideal way to >go would be to add to the resolver library and use the nsswitch.conf >mechanism. > >Well, I just discovered that the Samba project has at least started work in >that direction, with the winbind subproject. As far as I can tell it >currently only works for users and groups, but I doubt that adding >hostname resolution would be too difficult for somebody motivated.... > >http://us4.samba.org/samba/docs/man/winbind.html > > > From jep200305 at columbus.rr.com Mon Mar 1 09:52:30 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <40434606.4090902@whipple.cc> References: <200402292351.58732.rfunk@funknet.net> <40434606.4090902@whipple.cc> Message-ID: <20040301095230.23616294.jep200305@columbus.rr.com> Joe wrote: > I was under the impression samba already did this... If you have a Windows box that is a DHCP client with a netbios name of FOOBOX and it is not in any DNS server info and not in any /etc/hosts files, other Windows boxes can successfully do: ping foobox but Linux boxes can not. Linux boxes have to do a kludge like: nmblookup foobox then ping the address from the output of nmblookup. ping 192.168.0.108 The issue is to integrate SMB'ish name resolution into general name resolution, so that SMB'ish name resolution works for all commands, not just the SMB'ish commands. I'm thinking about the actual networks that novice users have of DHCP client Windows boxes with a cheap commodity firewall/NAT box. Name resolution is a weakness of adding a Linux box to such a network. Having a DNS server or WINS server or "logging into a domain" is not an option. Static IP is not an option. /etc/hosts does not work for DHCP clients. Pseudo-static IP by locking down DHCP clients to particular IP addresses is not an option. From blata at extent0006.entomology.ohio-state.edu Mon Mar 1 10:26:56 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get Message-ID: <40435640.6040503@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm brand new to debian. I am trying to use apt-get but I keep getting this error. What does it mean and how can I fix it? " Reading package lists....Error! E: Unable to write mmap - msync (5 input/output error) E: The package lists or status file could not be parsed or opened. " - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFAQ1Y/v+6+qSFv3TURAneDAJ4+MkgYZYJfj3oTgrzjlBnW9Oet9wCbBPoK lCXgzGkBdcxpKVzFDnIPPRU= =DTJD -----END PGP SIGNATURE----- From skippy at skippy.net Mon Mar 1 10:26:31 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040301095230.23616294.jep200305@columbus.rr.com> References: <200402292351.58732.rfunk@funknet.net><40434606.4090902@whipple.cc> <20040301095230.23616294.jep200305@columbus.rr.com> Message-ID: <1396.216.136.35.122.1078154791.squirrel@www.skippy.net> Jim P said: > The issue is to integrate SMB'ish name resolution into > general name resolution, so that SMB'ish name resolution > works for all commands, not just the SMB'ish commands. > > I'm thinking about the actual networks that novice users have > of DHCP client Windows boxes with a cheap commodity firewall/NAT box. > Name resolution is a weakness of adding a Linux box to such a network. I think it was Phil Hunter who mentioned Zeroconf: http://www.zeroconf.org/ Freshmeat had an announcement for a Zeroconf front-end called "Howl": http://www.porchdogsoft.com/products/howl/ Looking it over, that doesn't look like the magic bullet Jim's after. This may be closer: http://www.public.iastate.edu/~chadspen/homepage.html but probably still short of the mark. Rob Funk mentioned hacking the NIS resolver to make it WINS-aware. You could also hack dnsmasq or MaraDNS or any other DNS resolver to make them WINS-aware. From peter.king at utoronto.ca Mon Mar 1 10:52:21 2004 From: peter.king at utoronto.ca (Peter King) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Segfault suggestions? Message-ID: <20040301155221.GB23021@socrates> I've recently come into possession of a Dell Dimension v350, which is an old PII 350MHz computer with 128MB RAM. The problem is that it (apparently randomly) segfaults. So far it has crashed while running cdparanoia (but not cdrecord), apt-get, links, and a few other programs. The crashes are not reproducible: running the same program again usually works. The same phenomenon comes up under kernels 2.2.20, 2.4.24, and 2.6.2, running Debian Stable or Testing (depending). That makes it sound like hardware. So I ran 10 iterations of memtest overnight, but, after freezing the first time, it then turned up no memory errors at all. (Nor did it segfault, apparently.) So I guess it isn't the memory. Good thing, since I don't have any spare memory for that vintage machine lying around. So: If it isn't the software and it isn't the memory, what should I look at next? If the answer is "the CPU" -- how do I test that? I don't have a spare PII lying around... Any suggestions much appreciated. Thanks in advance! -- Peter King peter.king@utoronto.ca Department of Philosophy 215 Huron Street The University of Toronto (416)-978-3788 ofc Toronto, ON M5S 1A1 CANADA ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://www.colug.net/pipermail/colug/attachments/20040301/6dc00bad/attachment.bin From JBalint at alldata.net Mon Mar 1 11:00:51 2004 From: JBalint at alldata.net (Balint, Jess) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Segfault suggestions? Message-ID: <5789A42DD684734298BF9EC852D35111033AB24A@scad1exis02.alldata.net> If you want to get into it, see if there is a file called "core" sitting around after a program segfaults. If you have gdb then you can debug it. ~ > -----Original Message----- > From: colug-bounces@colug.net > [mailto:colug-bounces@colug.net]On Behalf > Of Peter King > Sent: Monday, March 01, 2004 10:52 AM > To: colug@colug.net > Subject: [COLUG] Segfault suggestions? > > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > From chris.fuhrman at tfcci.com Mon Mar 1 11:02:19 2004 From: chris.fuhrman at tfcci.com (Chris Fuhrman) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get In-Reply-To: <40435640.6040503@bugs.osu.edu> References: <40435640.6040503@bugs.osu.edu> Message-ID: <1078156939.22821.30.camel@icestorm.tfcc.com> What's the exact command you're running? On Mon, 2004-03-01 at 10:26, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I'm brand new to debian. I am trying to use apt-get but I keep getting > this error. What does it mean and how can I fix it? > > " > Reading package lists....Error! > E: Unable to write mmap - msync (5 input/output error) > E: The package lists or status file could not be parsed or opened. > " > > > - -- > Wade Pinkston > The Ohio State University > Extension Entomology > 1991 Kenny Rd > Columbus OH 43210 > phone: (614) 292-5274 > > Ipsa scientia potestas est > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.1 (GNU/Linux) > Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org > > iD8DBQFAQ1Y/v+6+qSFv3TURAneDAJ4+MkgYZYJfj3oTgrzjlBnW9Oet9wCbBPoK > lCXgzGkBdcxpKVzFDnIPPRU= > =DTJD > -----END PGP SIGNATURE----- > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -- Chris Fuhrman | Twenty First Century Communications chris.fuhrman@tfcci.com | Senior Software Engineer (W) 614-442-1215 x271 | (F) 614-442-5662 | PGP/GPG Public Key Available on Request From blata at extent0006.entomology.ohio-state.edu Mon Mar 1 11:11:43 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get In-Reply-To: <1078156939.22821.30.camel@icestorm.tfcc.com> References: <40435640.6040503@bugs.osu.edu> <1078156939.22821.30.camel@icestorm.tfcc.com> Message-ID: <404360BF.2080604@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 # apt-get install less Chris Fuhrman spewed verbiage on 03/01/2004 11:02 AM: | What's the exact command you're running? | | On Mon, 2004-03-01 at 10:26, Wade Pinkston wrote: | | I'm brand new to debian. I am trying to use apt-get but I keep getting | this error. What does it mean and how can I fix it? | | " | Reading package lists....Error! | E: Unable to write mmap - msync (5 input/output error) | E: The package lists or status file could not be parsed or opened. | " | | _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFAQ2C/v+6+qSFv3TURAnfKAJ9c2TDkN0TyCiE7GeFntQmVTa9xqQCfcqdC +tL4uk1kQKuOzz8JDCDVimw= =+AnB -----END PGP SIGNATURE----- From tom at functionalmedia.com Mon Mar 1 11:27:43 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Linux firewall In-Reply-To: <200402282208.i1SM8SfP003661@stones.wcbe.org> Message-ID: <5DBEEED0-6B9D-11D8-95CE-00039317745E@functionalmedia.com> I have heard good reports on ipcop.. I may take a look at it shortly. Sysadmin magazine just ran an article on it. Tom On Saturday, February 28, 2004, at 05:08 PM, firewall@stones.wcbe.org wrote: > Someone asked about firewalls at the meeting. > The two most common ones are Coyote Linux and LEAF/Bering. > > http://firewall.colug.net > > Coyote Linux is easier for novices to setup. > > LEAF/Bering is much more sophisticated and > also much more complicated to setup. > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From tom at functionalmedia.com Mon Mar 1 11:57:19 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get In-Reply-To: <40435640.6040503@bugs.osu.edu> Message-ID: <808F38A8-6BA1-11D8-95CE-00039317745E@functionalmedia.com> Are you running apt-get as root ?? Are you out of disk space ?? does /etc/apt/sources.list exist ?? Tom On Monday, March 1, 2004, at 10:26 AM, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I'm brand new to debian. I am trying to use apt-get but I keep getting > this error. What does it mean and how can I fix it? > > " > Reading package lists....Error! > E: Unable to write mmap - msync (5 input/output error) > E: The package lists or status file could not be parsed or opened. > " > > > - -- > Wade Pinkston > The Ohio State University > Extension Entomology > 1991 Kenny Rd > Columbus OH 43210 > phone: (614) 292-5274 > > Ipsa scientia potestas est > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.1 (GNU/Linux) > Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org > > iD8DBQFAQ1Y/v+6+qSFv3TURAneDAJ4+MkgYZYJfj3oTgrzjlBnW9Oet9wCbBPoK > lCXgzGkBdcxpKVzFDnIPPRU= > =DTJD > -----END PGP SIGNATURE----- > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From blata at extent0006.entomology.ohio-state.edu Mon Mar 1 12:07:38 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get In-Reply-To: <808F38A8-6BA1-11D8-95CE-00039317745E@functionalmedia.com> References: <808F38A8-6BA1-11D8-95CE-00039317745E@functionalmedia.com> Message-ID: <40436DDA.3000205@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 well first thanks for the pointers. turns out only my /boot portion is mounting and yes it is now 100% full. Now I can't edit the fstab file because there is no room left on disk. tom hanlon spewed verbiage on 03/01/2004 11:57 AM: | Are you running apt-get as root ?? | Are you out of disk space ?? | does /etc/apt/sources.list exist ?? | | Tom | | | | | On Monday, March 1, 2004, at 10:26 AM, Wade Pinkston wrote: | | I'm brand new to debian. I am trying to use apt-get but I keep getting | this error. What does it mean and how can I fix it? | | " | Reading package lists....Error! | E: Unable to write mmap - msync (5 input/output error) | E: The package lists or status file could not be parsed or opened. | " | | _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug |> |> | Tom Hanlon | Functional Media | 740-597-1472 | tom@functionalmedia.com | www.functionalmedia.com | _______________________________________________ | colug mailing list | colug@colug.net | http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFAQ23av+6+qSFv3TURAhW2AJ4iDg2XoZUJBzQ3w1/4LGRO10mFSgCeOPGs m6wg4rskxKDAgzVzFWxTCAA= =O7jz -----END PGP SIGNATURE----- From tom at functionalmedia.com Mon Mar 1 12:46:48 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] apt-get In-Reply-To: <40436DDA.3000205@bugs.osu.edu> Message-ID: <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> The old "disk is full" problem.. I encounter this all the time. Does anyone have a script that they run at startup or login, that checks disk space and delivers a message ?? Even my "user friendly" macintosh OSX laptop (BSD under the hood) behaves very badly when it runs out of space. The errors returned when the disk is full rarely say "disk is full" they usually say something like, "unable to write file" or some other less than specific error. Tom On Monday, March 1, 2004, at 12:07 PM, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > well first thanks for the pointers. > > turns out only my /boot portion is mounting and yes it is now 100% > full. Now I can't edit the fstab file because there is no room left on > disk. > > tom hanlon spewed verbiage on 03/01/2004 11:57 AM: > | Are you running apt-get as root ?? > | Are you out of disk space ?? > | does /etc/apt/sources.list exist ?? > | > | Tom > | > | > | > | > | On Monday, March 1, 2004, at 10:26 AM, Wade Pinkston wrote: > | > | I'm brand new to debian. I am trying to use apt-get but I keep > getting > | this error. What does it mean and how can I fix it? > | > | " > | Reading package lists....Error! > | E: Unable to write mmap - msync (5 input/output error) > | E: The package lists or status file could not be parsed or opened. > | " > | > | -- Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From michael_j_mangino at bankone.com Mon Mar 1 12:55:46 2004 From: michael_j_mangino at bankone.com (michael_j_mangino@bankone.com) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Free stuff Message-ID: I cleaned out my office this weekend and I have some stuff to get rid of. I have a Sony Trinitron 21" monitor with cabling that still has a decent picture. It isn't anywhere as bright as a modern LCD, but it still works wonderfully with both Linux and Windows. I also have a Decstation 5000/133 and a 5000/200 with a 19" monitor. I got this from a lister several years ago. If that person wants them back, let me know. I have no idea whether or not these work. I was going to work on the Linux port, but my MBA took up all of my time. Send me an email if you are interested. You should be prepared to pick these up at my house in Gahanna. Mike Mike Mangino, I&O Midrange Architecture 614-248-7664 W 614-306-0078 C 888-260-2391 P 614-428-7565 H michael_j_mangino@bankone.com This transmission may contain information that is privileged, confidential and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. From michael_j_mangino at bankone.com Mon Mar 1 13:22:56 2004 From: michael_j_mangino at bankone.com (michael_j_mangino@bankone.com) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] free items Message-ID: All items have been claimed. Mike Mike Mangino, I&O Midrange Architecture 614-248-7664 W 614-306-0078 C 888-260-2391 P 614-428-7565 H michael_j_mangino@bankone.com This transmission may contain information that is privileged, confidential and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. From chris at chrisbaker.net Mon Mar 1 15:24:30 2004 From: chris at chrisbaker.net (chris@chrisbaker.net) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed Message-ID: <404355AE.12999.183D97D@localhost> Hi, guys, I just thought I would check in to let everyone know that I have sort of reached a frustration point with a project that I am working on. I may consider paying someone else to do it. Basically, I need someone who can set up an Exim MTA on a Debian box. The Exim MTA will only be used as a "front door" to an Exchange server. I will add virus scanning and spam filtering on the box. I initially tried SuSE and abandoned that in favor of Debian. I am also going on a week's vacation starting on the evening of 12 March and definitely want this done before then. I imagine this would be a breeze for someone like Scott Sharkey or Phil Hunter. I'll be happy if I know half as much as those guys do. Anyway, if anyone knows of someone who can set up Exim on Debian with an anti-spam strategy called graylisting, please let me know. This is for the non-profit that I work for, but I may just pay for it myself. I feel like I have already taken way too long to get it done. Thanks for everything. From mccune at math.ohio-state.edu Mon Mar 1 16:00:17 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040301095230.23616294.jep200305@columbus.rr.com> References: <200402292351.58732.rfunk@funknet.net> <40434606.4090902@whipple.cc> <20040301095230.23616294.jep200305@columbus.rr.com> Message-ID: <20040301210017.GB11576@math.ohio-state.edu> On Mon, Mar 01, 2004 at 09:52:30AM -0500, Jim P enlightened us: > Joe wrote: > > > I was under the impression samba already did this... > > If you have a Windows box that is a DHCP client with a netbios > name of FOOBOX and it is not in any DNS server info and not in any > /etc/hosts files, other Windows boxes can successfully do: > > ping foobox > > but Linux boxes can not. Linux boxes have to do a kludge like: > > nmblookup foobox > > then ping the address from the output of nmblookup. > > ping 192.168.0.108 > At work and home, I hacked a script that works well with subshell expansions. The idea is that it "automates" the above two step process, so you can do something like this: ping `nmb foobox`; And it'll work fine. I use this frequently to rdesktop to all of our lab machines. The script is a one-liner: [silas@momar silas]$ cat `which nmb` #!/bin/bash # nmblookup "$1" | grep "$1<00>" | cut -d' ' -f1 Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/fe479323/attachment.bin From skippy at skippy.net Mon Mar 1 16:05:51 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:33 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <404355AE.12999.183D97D@localhost> References: <404355AE.12999.183D97D@localhost> Message-ID: <2925.216.136.35.122.1078175151.squirrel@www.skippy.net> chris@chrisbaker.net said: > Basically, I need someone who can set up an Exim MTA on a Debian > box. The Exim MTA will only be used as a "front door" to an Exchange > server. I will add virus scanning and spam filtering on the box. > > I initially tried SuSE and abandoned that in favor of Debian. I am also Why'd you abandon SuSE? Are you committed to Exim? Have you considered Postfix or qmail? What have you tried so far, and what have the results been? We might be able to help you help yourself! > going on a week's vacation starting on the evening of 12 March and > definitely want this done before then. So assuming you get it installed and operational tomorrow, that gives you less than two weeks to bang on it before you go off-site and (presumably) unavailable for a week? > I imagine this would be a breeze for someone like Scott Sharkey or > Phil Hunter. I'll be happy if I know half as much as those guys do. > Anyway, if anyone knows of someone who can set up Exim on Debian > with an anti-spam strategy called graylisting, please let me know. Must you use greylisting? It's still in "experimental" stages, if I read the description correctly. Some simple SMTP restrictions and SpamAssassin can do quite a bit to filter out the most egregious spam. The current trend to "confuse" SpamAssassin with lots and lots of random words has allowed more junk to trickle past my filters, but it's usually immediately obvious that it's spam. From pat at linuxcolumbus.com Mon Mar 1 16:08:59 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <404355AE.12999.183D97D@localhost> References: <404355AE.12999.183D97D@localhost> Message-ID: <20040301210859.GA11011@linuxcolumbus.com> On Mon, Mar 01, 2004 at 03:24:30PM -0500, chris@chrisbaker.net wrote: > Hi, guys, > I imagine this would be a breeze for someone like Scott Sharkey or > Phil Hunter. I'll be happy if I know half as much as those guys do. > Anyway, if anyone knows of someone who can set up Exim on Debian > with an anti-spam strategy called graylisting, please let me know. > Might I suggest http://www.qmailrocks.org/ (Just because it mentions debian install) Most of the instructions in this install can be skipped. and http://www.digitaleveryware.com/projects/greylisting/ (also has exim greylisting information) http://lists.puremagic.com/pipermail/greylist-users/2004-January/000400.html or http://www.jonatkins.com/qgreylist/ for greylisting. Setting up a forwarding gateway is extremely easy with qmail. echo ":theothermailserver_ip_or_name" > /var/qmail/smtproutes and making sure the proper daemons (qmail-smtpd,qmail-send) are running. Pat From rfunk at funknet.net Mon Mar 1 16:13:47 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] apt-get In-Reply-To: <40435640.6040503@bugs.osu.edu> References: <40435640.6040503@bugs.osu.edu> Message-ID: <200403011613.47122.rfunk@funknet.net> Wade Pinkston wrote: > I'm brand new to debian. I am trying to use apt-get but I keep getting > this error. What does it mean and how can I fix it? > > " > Reading package lists....Error! > E: Unable to write mmap - msync (5 input/output error) > E: The package lists or status file could not be parsed or opened. > " Wow. That's ugly. My best guess: Make sure you're running as root, and make sure you do "apt-get update" before doing anything else with apt-get. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From mccune at math.ohio-state.edu Mon Mar 1 16:12:07 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] apt-get In-Reply-To: <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> Message-ID: <20040301211207.GC11576@math.ohio-state.edu> On Mon, Mar 01, 2004 at 12:46:48PM -0500, tom hanlon enlightened us: > The old "disk is full" problem.. > > I encounter this all the time. Does anyone have a script that they run > at startup or login, that checks disk space and delivers a message ?? http://www.debian.org/releases/stable/i386/ch-partitioning.en.html This is why partitioning out /var (at the very least) is important. I usually have /, /boot, /usr, /var, /tmp, and /home on different partitions. /boot and /usr are less of a concern, as is /tmp if you use tmpfs or something akin to it. Basically any place where you write lots of data frequently shouldn't be part of the / file system in order to prevent system weirdness when the disk fills up. > Even my "user friendly" macintosh OSX laptop (BSD under the hood) > behaves very badly when it runs out of space. Unix likes some breathing room. Windows behaves weird too when it's out of disk. It's just more annoying about letting you know when you're getting "close" (I don't consider 25% free space on a 1gig partition "close" but that's just me.) > The errors returned when the disk is full rarely say "disk is full" > they usually say something like, "unable to write file" or some other > less than specific error. Well, this is because every piece of software doesn't bother to check *why* it couldn't write to disk. The list of things to check in this case would be rather long, and it'd be a pain for every software maintainer to perform each of these checks. The best advice I've been given regarding file system layout is that / should actually be mounted read-only on production servers. So should /usr, but I haven't met a system administrator who actually mounts /usr read only. /var should be the only place where data changes frequently (Hence /var's name (VARiable). Regards, -- Jeff McCune Network Administrator Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/09cd31ea/attachment.bin From rfunk at funknet.net Mon Mar 1 16:18:51 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Segfault suggestions? In-Reply-To: <20040301155221.GB23021@socrates> References: <20040301155221.GB23021@socrates> Message-ID: <200403011618.51649.rfunk@funknet.net> Peter King wrote: > That makes it sound like hardware. So I ran 10 iterations of memtest > overnight, but, after freezing the first time, it then turned up no > memory errors at all. (Nor did it segfault, apparently.) So I guess > it isn't the memory. Good thing, since I don't have any spare memory > for that vintage machine lying around. Well, it still could be the memory.... memtest can't try every possible scenario that could trigger a memory problem. But it's certainly time to test in other directions. > So: If it isn't the software and it isn't the memory, what should I > look at next? Could be the motherboard. I have a motherboard that gets flaky whenever memory is plugged into the second bank of SIMMs. (Thus limiting me to 256MB.) Try removing some of the memory to see if things are more reliable. > If the answer is "the CPU" -- how do I test that? I don't have a > spare PII lying around... Could be the CPU overheating. (Are you overclocking?) One often-used test is a kernel compile. Google for "SIG11".... Swapping components is always a good way of determining the problem, as long as you can fairly reliably reproduce the problem. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From mccune at math.ohio-state.edu Mon Mar 1 16:16:36 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Linux firewall In-Reply-To: <5DBEEED0-6B9D-11D8-95CE-00039317745E@functionalmedia.com> References: <200402282208.i1SM8SfP003661@stones.wcbe.org> <5DBEEED0-6B9D-11D8-95CE-00039317745E@functionalmedia.com> Message-ID: <20040301211636.GD11576@math.ohio-state.edu> On Mon, Mar 01, 2004 at 11:27:43AM -0500, tom hanlon enlightened us: > I have heard good reports on ipcop.. > > I may take a look at it shortly. Sysadmin magazine just ran an article > on it. > > Tom > If you have the time, I strongly encourage people to roll their own netfilter iptables scripts to setup their firewall. It was the single most educational experience in low-level networking I've received. My firewall script is pretty convoluted and I've got a ton of "extras" built in, but if you'd like to look at it as an example, please let me know. It's easy to open ports, adjust protocol priorities, and shape traffic with it. I've got a bunch of leechers in my apartment, so traffic shaping is essential for us, but I find most people aren't all that interested in it. Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/7c0520ab/attachment.bin From pat at linuxcolumbus.com Mon Mar 1 16:19:08 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040301210859.GA11011@linuxcolumbus.com> References: <404355AE.12999.183D97D@localhost> <20040301210859.GA11011@linuxcolumbus.com> Message-ID: <20040301211908.GB11011@linuxcolumbus.com> On Mon, Mar 01, 2004 at 04:08:59PM -0500, pat@linuxcolumbus.com wrote: > Setting up a forwarding gateway is extremely easy with qmail. echo > ":theothermailserver_ip_or_name" > /var/qmail/smtproutes and making sure > the proper daemons (qmail-smtpd,qmail-send) are running. > Should be /var/qmail/control/smtproutes. Pat From chris at chrisbaker.net Mon Mar 1 16:30:30 2004 From: chris at chrisbaker.net (chris@chrisbaker.net) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <2925.216.136.35.122.1078175151.squirrel@www.skippy.net> References: <404355AE.12999.183D97D@localhost> Message-ID: <40436526.27472.1C045DF@localhost> > Why'd you abandon SuSE? It seemed to have too many unsolvalbe problems. I still have that box sitting here. It also installed a bunch of stuff I didn't need and crawled on a 350 CPU. > Are you committed to Exim? Have you considered Postfix or qmail? I may try Qmail, but haven't heard much about graylisting with Qmail. I have heard that Postfix doesn't work well at all with it. I did already spend $75 on an Exim book. > What have you tried so far, and what have the results been? We might be > able to help you help yourself! They have been when I run the make command to compile Exim. It keeps saying that it can't find this or that. It finds one thing and then can't find something else. > So assuming you get it installed and operational tomorrow, that gives you > less than two weeks to bang on it before you go off-site and (presumably) > unavailable for a week? I just don't want to drag this out any longer. > Must you use greylisting? It's still in "experimental" stages, if I read > the description correctly. It doesn't seem that experimental. It sounds like it works pretty well for the ones who are doing it. I would love to be able to put on my resume: "I solved my employer's spam problem." Chris From chris at chrisbaker.net Mon Mar 1 16:30:30 2004 From: chris at chrisbaker.net (chris@chrisbaker.net) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040301210859.GA11011@linuxcolumbus.com> References: <404355AE.12999.183D97D@localhost> Message-ID: <40436526.18395.1C04504@localhost> > http://www.qmailrocks.org/ (Just because it mentions debian install) Most > of the instructions in this install can be skipped. I'll check that out, but I did already spend about $75 for an Exim book. A lot of good that book has done. > and > http://www.digitaleveryware.com/projects/greylisting/ (also has exim > greylisting information) > http://lists.puremagic.com/pipermail/greylist-users/2004-January/000400.html > or > http://www.jonatkins.com/qgreylist/ I have read the first two, and both had incomplete instructions that I have either followed incorrectly or were incorrect or incomplete to begin with. I am also on the Exim discussion group. I have not checked out the third. > Setting up a forwarding gateway is extremely easy with qmail. echo > ":theothermailserver_ip_or_name" > /var/qmail/smtproutes and making sure > the proper daemons (qmail-smtpd,qmail-send) are running. Everybody said that Exim was easy, too. Either they lied or they have some type of magical ability that I don't have, even though my IQ (135) test said that I am lot smarter than most. Chris From mccune at math.ohio-state.edu Mon Mar 1 16:43:42 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <404355AE.12999.183D97D@localhost> References: <404355AE.12999.183D97D@localhost> Message-ID: <20040301214342.GE11576@math.ohio-state.edu> On Mon, Mar 01, 2004 at 03:24:30PM -0500, chris@chrisbaker.net enlightened us: > Hi, guys, > > I just thought I would check in to let everyone know that I have sort of > reached a frustration point with a project that I am working on. I may > consider paying someone else to do it. > > Basically, I need someone who can set up an Exim MTA on a Debian > box. The Exim MTA will only be used as a "front door" to an Exchange > server. I will add virus scanning and spam filtering on the box. I've setup an Exim4 / MailScanner (ClamAV+Mcafee) / SpamAssassin running on debian woody at Linguistics. The server works great, and I *strongly* encourage you to stick with Exim, as it's *great* to work with it's transport and router filters. I can't explicitely work on the project, but I'd be happy to help with any questions you might have, or provide any example config files from our setup. We run 2 Exim4 processes on the box. One "incoming" process, one "outgoing" process. MailScanner sits in between these two processes and is the only way that they can exchange messages. The idea is that an email comes into the incoming process, and it accepts / rejects it. If it accepts, it dumps it into the waiting queue which MailScanner checks every 60 seconds or so. Once MailScanner picks it up, it scans it for viruses using ClamAV (updated twice daily) and then hands it off to the outgoing exim process, which will either relay it to the outside world, or deliver it to the user's mailbox. Local delivery is handled by procmail, which first sends the message through the spamd process for site-wide spam filtering, retaining user preferences. Everything from the user perspective is configured through a web based interface which stores all settings in a MySQL database. exim4 even allowed me to easily move the ~/.forward file into SQL. > I initially tried SuSE and abandoned that in favor of Debian. I am also > going on a week's vacation starting on the evening of 12 March and > definitely want this done before then. > > I imagine this would be a breeze for someone like Scott Sharkey or > Phil Hunter. I'll be happy if I know half as much as those guys do. > Anyway, if anyone knows of someone who can set up Exim on Debian > with an anti-spam strategy called graylisting, please let me know. I haven't heard about greylisting. Any links? Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/dd590ed9/attachment.bin From mccune at math.ohio-state.edu Mon Mar 1 16:50:17 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <40436526.27472.1C045DF@localhost> References: <404355AE.12999.183D97D@localhost> <40436526.27472.1C045DF@localhost> Message-ID: <20040301215017.GF11576@math.ohio-state.edu> On Mon, Mar 01, 2004 at 04:30:30PM -0500, chris@chrisbaker.net enlightened us: > > Why'd you abandon SuSE? > > It seemed to have too many unsolvalbe problems. I still have that box > sitting here. It also installed a bunch of stuff I didn't need and crawled > on a 350 CPU. > > > Are you committed to Exim? Have you considered Postfix or qmail? > > I may try Qmail, but haven't heard much about graylisting with Qmail. I > have heard that Postfix doesn't work well at all with it. I did already > spend $75 on an Exim book. > > > What have you tried so far, and what have the results been? We might be > > able to help you help yourself! > > They have been when I run the make command to compile Exim. It > keeps saying that it can't find this or that. It finds one thing and then > can't find something else. > Whoa whoa whoa... Why are you compiling exim yourself? There's exim3 and exim4 packages for debian. exim4 packages aren't a standard part of woody, but you can get backported versions of them easy enough. I *strongly* discourage installing software outside of the dpkg system on debian. It sorta defeats much of what makes debian so easy to administer. In the entire setup, you should be able to use "normal" Debian packages. I think the only software that I use which wasn't packaged for debian was MailScanner, and that installs nicely into it's own little section of the file system. > > So assuming you get it installed and operational tomorrow, that gives you > > less than two weeks to bang on it before you go off-site and (presumably) > > unavailable for a week? > > I just don't want to drag this out any longer. > > > Must you use greylisting? It's still in "experimental" stages, if I read > > the description correctly. > > It doesn't seem that experimental. It sounds like it works pretty well for > the ones who are doing it. > > I would love to be able to put on my resume: "I solved my employer's > spam problem." > Our system, as I described previously, works almost 100% (I still see a spam message every other day or so, but it's only 1, and bayes eventually learns them). I haven't even setup a site-wide bayes database yet, which would get me even closer to perfection. I would start with a "decent" implementation rather than trying to go all out on the first try. Once you get spamassassin and virus scanning working, I'd go and then try to integrate other features into the system. Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/2f3b96f2/attachment.bin From skippy at skippy.net Mon Mar 1 16:52:08 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <40436526.27472.1C045DF@localhost> References: <404355AE.12999.183D97D@localhost> <40436526.27472.1C045DF@localhost> Message-ID: <3072.216.136.35.122.1078177928.squirrel@www.skippy.net> chris@chrisbaker.net said: >> Are you committed to Exim? Have you considered Postfix or qmail? > > I may try Qmail, but haven't heard much about graylisting with Qmail. I > have heard that Postfix doesn't work well at all with it. I did already > spend $75 on an Exim book. Postfix has a greylisting plug-in. It's perl-based, though, so it's not exactly optimized for high-volume (what with the expense of spawning the perl interpreter). Until they daemonize it, I'm not keen on it. >> What have you tried so far, and what have the results been? We might be >> able to help you help yourself! > > They have been when I run the make command to compile Exim. It > keeps saying that it can't find this or that. It finds one thing and then > can't find something else. Ah ha! You're not "Doing it the Debian Way(tm)"! ;) # apt-get install exim Exim is Debian's default MTA. Why do you need to compile from source? What about the stock config is lacking? >> Must you use greylisting? It's still in "experimental" stages, if I >> read >> the description correctly. > > It doesn't seem that experimental. It sounds like it works pretty well for > the ones who are doing it. I looked at it when Josh Glover shared the link. I googled a bit more. The paper Josh linked was a "proof of concept" implementation, as I recall. Though certainly there are people running experimental services all day long on production servers -- I tend to prefer time-proven solutions for dealing with things like _other people's email_. > I would love to be able to put on my resume: "I solved my employer's > spam problem." "Mitigated" or "minimized" is probably the best you'll ever get; unless you implement a mandatory challenge-response (a la DSPAM), which is likely to piss off your users when they buy airline tickets and the automated confirmation email fails a challenge. From rfunk at funknet.net Mon Mar 1 17:02:36 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <40436526.27472.1C045DF@localhost> References: <404355AE.12999.183D97D@localhost> <40436526.27472.1C045DF@localhost> Message-ID: <200403011702.36512.rfunk@funknet.net> chris@chrisbaker.net wrote: > I may try Qmail, but haven't heard much about graylisting with Qmail. I > have heard that Postfix doesn't work well at all with it. I did already > spend $75 on an Exim book. Postfix and Qmail are both more secure than Exim, which in turn is more secure than sendmail. If you're connecting to the internet, security needs to be a criterion. But both Postfix and Exim are more Debian-friendly and package-friendly than Qmail. > They have been when I run the make command to compile Exim. It > keeps saying that it can't find this or that. It finds one thing and > then can't find something else. Why are you building from source instead of "apt-get install exim" (or exim-tls)? Debian's install should get you started with configuration too. Postfix is also apt-gettable. > > So assuming you get it installed and operational tomorrow, that gives > > you less than two weeks to bang on it before you go off-site and > > (presumably) unavailable for a week? > > I just don't want to drag this out any longer. You *really* don't want to leave town right after setting up something like this. At least, not if you don't want a lynch mob waiting for you when you come back. Wait until it's been running stable for a while before going on vacation. Or, wait to activate this until after your vacation. > > Must you use greylisting? It's still in "experimental" stages, if I > > read the description correctly. > > It doesn't seem that experimental. It sounds like it works pretty well > for the ones who are doing it. But not that many people are doing it. Thus it's still experimental. Spamassassin is not experimental. Graylisting is. > I would love to be able to put on my resume: "I solved my employer's > spam problem." You can't do that if you hire someone else to set it up. :-) -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From mccune at math.ohio-state.edu Mon Mar 1 17:04:47 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040301214342.GE11576@math.ohio-state.edu> References: <404355AE.12999.183D97D@localhost> <20040301214342.GE11576@math.ohio-state.edu> Message-ID: <20040301220447.GG11576@math.ohio-state.edu> I should note that the only reason I decided to go with backported exim4 debian packages on woody rather than the default exim package that comes with the distribution was because exim4 can pull any data that would normally be in a flat file from a mysql database. exim3 would have worked fine if it weren't for this feature, which in retrospect wasn't nearly as important as I thought it was. Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/b7b08dd2/attachment.bin From jacob at teched.net Mon Mar 1 17:43:29 2004 From: jacob at teched.net (J. Jacob Hopkins) Date: Sat Jan 8 01:34:34 2005 Subject: [COLUG] JTAG/IEEE 1149.1 Message-ID: <20040301224329.GF20913@noroute.teched.net> Does anyone have sharable experiences with Free/Open JTAG/IEEE 1149.1 software? Jacob jacob@teched.net From mccune at math.ohio-state.edu Mon Mar 1 17:44:47 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] SprintPCS+Knoppix=SWEET Message-ID: <20040301224447.GB16496@math.ohio-state.edu> For one of the project teams I'm involved in, we're traveling to Richmond, VA for competition for 5 days. In preperation, I decided to solder the USB cable that connects my Sanyo SCP-4900 cell phone to my Linux notebook in order to have the 5v USB line charge the phone as well. It was pretty painless to do, though I would recommend people just buy cables that already do this, as they're only $15 online now. http://www.natecarlson.com/linux/sanyo-4900.php I decided to start clean with debian on my notebook to minimize any ill-effects that might come up at competition, so I did a debootstrap from Knoppix. Much to my surpise, knoppix already has a complete configuration for sprinpcs included on the CD. All I had to do was plug the phone into the USB port, link /dev/modem to /dev/ttyACM0 and then execute pon sprint. That's pretty sweet. With the charging cable, I can now stay online with my notebook for the entire 8 hour bus ride, as well as the entire day in the pit area in the arena, so I'm pretty happy. I tested this by leaving the Connection up all of last night. Worked fine, as long as I stayed sshed somewhere with the KeepAliveProtocols option turned on. Otherwise, it seems to automatically shut off after 30 minutes. Anyway, if you have SprintPCS and travel a lot, it might be worth your while to investigate which phones are detected as standard USB acm modems under Linux. It seems that if you carry the $15 cable and a knoppix CD, you can easily get on the internet anywhere you have PCS signal. It'll be interesting for the 8 hour bus ride and the hotel, since everyone else has powerbooks... They'll "automatically" get a DHCP address from my prism2 card acting as an access point and will be routed through the cell phone. Wireless is pretty sweet. Oh, and if it wasn't clear, if you have a Vision plan with sprint, ppp internet access from a notebook works just fine, and is completely free (It's considered Vision usage), but it's explicitely forbidden in the contract you sign. However, I've talked to 3 different sprint representatives (One in Akron, one in Columbus, one on the phone) and they all state that Sprint knows people are doing it, and doesn't mind as long as you don't use it heavily. One of them said that it was a marketing decision to allow this fact to spread on the net without sprint directly marketing it as unmetered cellular internet for notebooks. Regardless, it's not an "official" feature, so if you're thinking of doing this for business, or are going to rely on it being there, it may go away in the future. I've been using it around the holdiays when traveling since last year. It's great. Around columbus, I get 10-15K/sec (150kbit), and up in north east ohio I get a solid 200kbit with a steady connection from Akron to Ashtabula. Getting closer and closer to my dream of high speed palmtop connectivity... Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/56fbfcf1/attachment.bin From colug at cencula.com Mon Mar 1 17:50:40 2004 From: colug at cencula.com (Michael D. Cencula) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] DesqView /X Message-ID: <200403011750.40710.colug@cencula.com> Does anyone know where I might be able to obtain a copy of this vintage software? I've been checking eBay, but nobody seems to be selling any copies right now. Thanks, Mike From jep200305 at columbus.rr.com Mon Mar 1 18:13:40 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] DesqView /X In-Reply-To: <200403011750.40710.colug@cencula.com> References: <200403011750.40710.colug@cencula.com> Message-ID: <20040301181340.05e10393.jep200305@columbus.rr.com> "Michael D. Cencula" wrote: > Does anyone know where I might be able to obtain a copy of this vintage > software? Wow, I had to google for that. http://en.wikipedia.org/wiki/DESQview What aspect of Desqview is it that you need? X terminal? multitasking DOS? running Windows 3.* in an X terminal? Considering the nature of this list, would you consider stuff like dosemu or cygwin? From colug at cencula.com Mon Mar 1 18:37:50 2004 From: colug at cencula.com (Michael D. Cencula) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] DesqView /X In-Reply-To: <20040301181340.05e10393.jep200305@columbus.rr.com> References: <200403011750.40710.colug@cencula.com> <20040301181340.05e10393.jep200305@columbus.rr.com> Message-ID: <200403011837.50998.colug@cencula.com> On Monday 01 March 2004 6:13 pm, Jim P wrote: > "Michael D. Cencula" wrote: > > Does anyone know where I might be able to obtain a copy of this vintage > > software? > > Wow, I had to google for that. http://en.wikipedia.org/wiki/DESQview > > What aspect of Desqview is it that you need? X terminal? > multitasking DOS? running Windows 3.* in an X terminal? > > Considering the nature of this list, > would you consider stuff like dosemu or cygwin? Well, I'm trying to set up a DOS box (freedos) with a NIC card that can run headless. I've got a single board computer that's plugged into one slot of a a 4 slot ISA passive backplane. It's got a nice compact case (5" x 8" x 10"), power supply, and 4MB flash disk. Essentially, I want to use it as a dedicated machine to run my universal programmer. The programmer has an 8-bit ISA card that plugs into the host computer. I've got a 3com 3c509 that will occupy the third slot, leaving the 4th slot unused. In the past, I've kept an old 486 around just for running the programmer, but I don't like having to keep a full-sized PC available for the few times I need to do chip programming. The programmer software is dos-based. Essentially, I want to be able to operate the the programmer and associated software from a remote (linux) machine. I'm currently playing around with telnetd off simtel, but it's pretty buggy and would like a solution that handles block characters better. The coolest solution so far seems to be DesqView /X, but one other possibility that looks promising is "Everywhere Access". Anyone have other suggestions? From jep200305 at columbus.rr.com Mon Mar 1 19:03:24 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <200403011837.50998.colug@cencula.com> References: <200403011750.40710.colug@cencula.com> <20040301181340.05e10393.jep200305@columbus.rr.com> <200403011837.50998.colug@cencula.com> Message-ID: <20040301190324.21bdbbd8.jep200305@columbus.rr.com> Michael D. Cencula wrote: > Well, I'm trying to set up a DOS box (freedos) with a NIC card that can run > headless. ... > I want to use it as a dedicated machine to run my universal programmer. ... > I want to be able to operate the the programmer and associated > software from a remote (linux) machine. Gotcha. > Anyone have other suggestions? I don't know about remote control of DOS over ethernet. I wonder if they exists a command like "ctty eth0". Hmmm, instead of remote operation over a(n ethernet) network, how about using ctty command for remote operation over RS-232? How well behaved is your program? (Many TUI programs that blasted to the screen don't work good over serial ports.) I know RS-232 seems crude, but you _are_ trying to do legacy stuff with legacy hardware on a legacy OS. > The programmer software is dos-based. I understand. Most were and most are still MS based. Will it run on Windows? If so, then VNC would be an option, although hard to automate (like from make). Can you automate the DOS program? Can you run it by feeding its standard input from a file? > The programmer has an > 8-bit ISA card that plugs into the host computer. Needham's PB-10? Of course, it'd be really cool if you could figure out some open-source driver for it. I fully realize how big an effort that would be. From jonadab at bright.net Mon Mar 1 19:07:24 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] apt-get In-Reply-To: <20040301211207.GC11576@math.ohio-state.edu> References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> <20040301211207.GC11576@math.ohio-state.edu> Message-ID: Jeff McCune writes: > The best advice I've been given regarding file system layout is that > / should actually be mounted read-only on production servers. So > should /usr, but I haven't met a system administrator who actually > mounts /usr read only. Mounting /usr read-only is an interesting idea, but unless I'm missing something it effectively means a reboot any time you have to change anything on /usr. There are circumstances where that would be acceptable, but for most purposes it would be a royal pain, I think. > /var should be the only place where data changes frequently (Hence > /var's name (VARiable). Ah, so _that's_ why /var is called that. I always wondered why it wasn't /log or somesuch, and why /var/www and /var/spool are there. That explains it. /home might also contain info that changes pretty regularly. For that matter, I find that on an actively-used system some of the stuff in /etc changes more often than would perhaps be expected. (On a desktop system, of course, /etc changes regularly and /home continuously.) -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From colug at cencula.com Mon Mar 1 19:37:31 2004 From: colug at cencula.com (Michael D. Cencula) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <20040301190324.21bdbbd8.jep200305@columbus.rr.com> References: <200403011750.40710.colug@cencula.com> <200403011837.50998.colug@cencula.com> <20040301190324.21bdbbd8.jep200305@columbus.rr.com> Message-ID: <200403011937.31696.colug@cencula.com> On Monday 01 March 2004 7:03 pm, Jim P wrote: > Michael D. Cencula wrote: > > Well, I'm trying to set up a DOS box (freedos) with a NIC card that can > > run headless. > > I want to be able to operate the the programmer and associated > > software from a remote (linux) machine. > > Gotcha. > > > Anyone have other suggestions? > > The programmer software is dos-based. > > I understand. Most were and most are still MS based. > Will it run on Windows? If so, then VNC would be an option, > although hard to automate (like from make). > > Can you automate the DOS program? > Can you run it by feeding its standard input from a file? Interesting idea. It may work for some features, but for other functions of the device (like testing 74xx series chips) I would need to see the output of the program. > > The programmer has an > > 8-bit ISA card that plugs into the host computer. > > Needham's PB-10? Actually, it's an AllMax by eetools (www.eetools.com) http://www.marvac.com/products/eetools/allmax_tech.htm > > Of course, it'd be really cool if you could figure > out some open-source driver for it. I fully realize > how big an effort that would be. Yikes. Although this is definitely doable, it's probably more effort than I'm willing to do right now. ;-) I haven't ruled out the RS232 concept. I'm aware of a program (also off simtel) called kbfake that's supposed to redirect keyboard and screen output to a com port. I haven't had a chance to check it out yet, though. From blata at extent0006.entomology.ohio-state.edu Mon Mar 1 19:43:21 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] win_gnupg Message-ID: <4043D8A9.2090203@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 anybody get gnupg to work under winblows with thunderbird? I created C:\program files\gnuPG\ put all of the gnupg files in to it. then I created C:\gnupg . At that point I ran thunderbird with enigmail installed and pointing at c:\program files\gnuPG but I get nothing when it comes to anything regarding gnupg. - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAQ9ipv+6+qSFv3TURAlTlAJ4vtl837Wfp++b5ZiwmcA4elsgZkQCgiPY2 Xju+qP9Vka7OGEcQtdLh/p0= =uZSE -----END PGP SIGNATURE----- From skippy at skippy.net Mon Mar 1 19:52:45 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] apt-get In-Reply-To: References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> <20040301211207.GC11576@math.ohio-state.edu> Message-ID: <4043DADD.7030901@skippy.net> Jonadab the Unsightly One wrote: > Jeff McCune writes: > > >>The best advice I've been given regarding file system layout is that >>/ should actually be mounted read-only on production servers. So >>should /usr, but I haven't met a system administrator who actually >>mounts /usr read only. > > > Mounting /usr read-only is an interesting idea, but unless I'm missing > something it effectively means a reboot any time you have to change > anything on /usr. There are circumstances where that would be > acceptable, but for most purposes it would be a royal pain, I think. See the mount manpage. -o Options remount "Attempt to remount an already-mounted file system. This is commonly used to change the mount flags for a file system, especially to make a readonly file system write- able. It does not change device or mount point." You could remount the read-only mount as read/write, then remount it read-only again after changes are applied. From blata at extent0006.entomology.ohio-state.edu Mon Mar 1 19:48:08 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] win_gnupg In-Reply-To: <4043D8A9.2090203@bugs.osu.edu> References: <4043D8A9.2090203@bugs.osu.edu> Message-ID: <4043D9C8.4090801@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm an idiot. I went over everything deleted thunderbird reinstalled every thing. and now it works just fine. Sorry. You wrote this to me on 03/01/2004 07:43 PM: | anybody get gnupg to work under winblows with thunderbird? I created | C:\program files\gnuPG\ put all of the gnupg files in to it. then I | created C:\gnupg . At that point I ran thunderbird with enigmail | installed and pointing at c:\program files\gnuPG but I get nothing | when it comes to anything regarding gnupg. | | -- | Wade Pinkston | The Ohio State University | Extension Entomology | 1991 Kenny Rd | Columbus OH 43210 | phone: (614) 292-5274 | | Ipsa scientia potestas est | | Windows,a | 32 bit graphical interface for a | 16 bit patch to an | 8 bit operating system internally coded for a | 4 bit processor written by a | 2 bit company that can't stand | 1 bit of competition | | GnuPG Key ID 0x216FDD35 | gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAQ9nIv+6+qSFv3TURAgSpAKCPrZAr7XDRtcBx6BGgzFP73qqjJQCfZIPZ eN+Z8vc0erRt5Y2g7AGFf4M= =eZ8u -----END PGP SIGNATURE----- From skippy at skippy.net Mon Mar 1 19:54:41 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <200403011937.31696.colug@cencula.com> References: <200403011750.40710.colug@cencula.com> <200403011837.50998.colug@cencula.com> <20040301190324.21bdbbd8.jep200305@columbus.rr.com> <200403011937.31696.colug@cencula.com> Message-ID: <4043DB51.9060302@skippy.net> Michael D. Cencula wrote: > I haven't ruled out the RS232 concept. I'm aware of a program (also off > simtel) called kbfake that's supposed to redirect keyboard and screen output > to a com port. I haven't had a chance to check it out yet, though. I've never used them, and have always wondered how well it would work, but there do exist KVM switches that operate over TCP/IP. I wonder if that might be another possible solution. From jep200305 at columbus.rr.com Mon Mar 1 20:34:52 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <200403011937.31696.colug@cencula.com> References: <200403011750.40710.colug@cencula.com> <200403011837.50998.colug@cencula.com> <20040301190324.21bdbbd8.jep200305@columbus.rr.com> <200403011937.31696.colug@cencula.com> Message-ID: <20040301203452.488facb2.jep200305@columbus.rr.com> Michael D. Cencula wrote: > On Monday 01 March 2004 7:03 pm, Jim P wrote: > > Can you automate the DOS program? > > Can you run it by feeding its standard input from a file? > > Interesting idea. It may work for some features, but for other functions of > the device (like testing 74xx series chips) I would need to see the output of > the program. Feeding the standard input from a file is an issue of automation, not remote operation, although what helps automation might aid remote operation. As far as viewing output, how about redirecting output to a file that a later program could display or parse for status? It'd be nice to have a program (perhaps script) that returns a pass/fail result that would play well with being invoked from make. I call my EPROM programmer program from a batch file invoked from make. I see the results because they are still on the (not remote) screen. Again, the issue of well-behaved programs arises. If the output is written directly to video memory, redirecting output to a file won't help. Same issues about direct video memory access for redirecting I/O with ctty. > > Of course, it'd be really cool if you could figure > > out some open-source driver for it. I fully realize > > how big an effort that would be. > > Yikes. Although this is definitely doable, it's probably more effort than I'm > willing to do right now. ;-) > > I haven't ruled out the RS232 concept. I'm aware of a program (also off > simtel) called kbfake that's supposed to redirect keyboard and screen output > to a com port. If it can catch writes to video memory, that's pretty good. Programs like Carbon Copy or PC Anywhere could. There were network versions also. Telix or Procomm might be able to be used in some kind of reverse manner to allow remote control of a DOS box over RS-232. Google for ctty Remote operation of a DOS box over ethernet is such an obvious thing for folks to do. I wonder if there is an OSS text version of VNC. This would not help automation. Another crude approach, is for the DOS box to run a script that looks for a new file on a share. The Linux box would put files for the work to do in a directory. When the DOS box detected the new files, it would do the work described in them and save the results on the share. This should play well with networks and automation, although saving the output could still be difficult if your programmer program writes directly to video memory. One way or another, you'll get remote operation working. From jep200305 at columbus.rr.com Mon Mar 1 21:55:19 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <200402292351.58732.rfunk@funknet.net> References: <200402292351.58732.rfunk@funknet.net> Message-ID: <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Thanks Rob for remembering this issue and continuing to think about it. Thanks to Scott, Joe and Jeff for trying to help. For myself, I just cope by using nmblookup, and Jeff's hack helps. The serious part though, is my consideration of ordinary non-technical users. For their sake, the solution will have to something incorporated in a distro whenever Samba is being used, or maybe even when Samba is not being used. It must require no effort on their part. Rob's technique of adding to the resolver library sounds the most attractive to me, although it's way over my head to either judge or attempt. Perhaps the Samba folks are the ones to provoke an itch among, which scratched would propagate to all the distros. Or perhaps provoking an itch amongst someone of a distro who would push some patches toward Samba and pull the result back for seamless integration into a distro. Who else should this itch be propagated to? From jep200305 at columbus.rr.com Mon Mar 1 23:04:30 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <4043DB51.9060302@skippy.net> References: <200403011750.40710.colug@cencula.com> <200403011837.50998.colug@cencula.com> <20040301190324.21bdbbd8.jep200305@columbus.rr.com> <200403011937.31696.colug@cencula.com> <4043DB51.9060302@skippy.net> Message-ID: <20040301230430.3ec6cb58.jep200305@columbus.rr.com> Scott wrote: > Michael D. Cencula wrote: > > I haven't ruled out the RS232 concept. I'm aware of a program (also off > > simtel) called kbfake that's supposed to redirect keyboard and screen output > > to a com port. I haven't had a chance to check it out yet, though. > > I've never used them, and have always wondered how well it would work, > but there do exist KVM switches that operate over TCP/IP. I wonder if > that might be another possible solution. Technically, it would certainly solve the remote control, but we'll have to hear from Michael about if it's affordable. A TCP/IP KVM switch would be awkward for transferring the files to burn, but plain ole' SMB on the regular ethernet port would be fine for that. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Yet another completely different approach, would be to run a PC emulator on an ISA Linux box and configure the emulator so that access to the programmer's I/O ports is passed to actual hardware, instead of being faked. This might actually work. One would certainly have to read the emulator manuals to see how to configure it for actual access to real hardware. Check out VMware and Bochs http://bochs.sourceforge.net/ A virtual machine like Plex86 http://plex86.sourceforge.net/ might also work, _fast_. These techniques might be easier than figuring out how to remotely use a DOS box. You would likely still use a separate box just for the ISA bus. From mccune at math.ohio-state.edu Mon Mar 1 23:22:50 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:35 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Message-ID: <20040302042250.GC22453@math.ohio-state.edu> On Mon, Mar 01, 2004 at 09:55:19PM -0500, Jim P enlightened us: > Thanks Rob for remembering this issue and continuing to think about it. > Thanks to Scott, Joe and Jeff for trying to help. > > For myself, I just cope by using nmblookup, and Jeff's hack helps. > > The serious part though, is my consideration of ordinary non-technical > users. For their sake, the solution will have to something incorporated > in a distro whenever Samba is being used, or maybe even when Samba is not > being used. It must require no effort on their part. > > Rob's technique of adding to the resolver library sounds the most > attractive to me, although it's way over my head to either judge > or attempt. > > Perhaps the Samba folks are the ones to provoke an itch among, > which scratched would propagate to all the distros. Or perhaps > provoking an itch amongst someone of a distro who would push some > patches toward Samba and pull the result back for seamless > integration into a distro. > > Who else should this itch be propagated to? > Nmblookup has always bugged me a bit, but I never use it enough to justify the time spent fixing it. I'm not sure if anyone else has mentioned this, but dynamic DNS might be a good solution. It can be implemented with already existing tools as well. The idea is that you have your dhcpd server take the hostname the client supplies, and then hands that off to the dns server which will make an entry for that hostname and the IP address that dhcpd assigned it. Not really a solution on a non-dhcp network, but it would allow you to ping foobox using dns, which is clean. All existing tools would then work, since the name can be "properly" resolved. I believe dhcpd and bind support this. It's not a bad idea to run internal DNS anyway, which I do at my apartment. I'm sure there's howto's out there on the subject. Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/e60b5e31/attachment.bin From mccune at math.ohio-state.edu Mon Mar 1 23:27:21 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] apt-get In-Reply-To: <4043DADD.7030901@skippy.net> References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> <20040301211207.GC11576@math.ohio-state.edu> <4043DADD.7030901@skippy.net> Message-ID: <20040302042721.GD22453@math.ohio-state.edu> mount -o remount,rw /usr && apt-get update \ && apt-get install foobar ; mount -o remount,ro /usr If you find yourself rebooting to fix something, clear your mind of windows, and then ask yourself if your ideal operating system would require you to reboot for this task. Chances are that answer will match the behavior of linux. With that said, Linux isn't my ideal OS, but it's the closest I've seen to date. =) Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040301/27776cd7/attachment.bin From gate at ilive4code.net Tue Mar 2 00:14:11 2004 From: gate at ilive4code.net (Greg Sidelinger) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] SprintPCS+Knoppix=SWEET In-Reply-To: <20040301224447.GB16496@math.ohio-state.edu> References: <20040301224447.GB16496@math.ohio-state.edu> Message-ID: <1078204451.4640.31.camel@sanitarium> Just want to say that this is a pretty good deal if you have sprint. I have a samsung n400 and have been using this same setup for over a year. I was even playing Never Winter Nights online with it once well on the interstate and it was even playable. I don't recommend downloading a new kernel over it but it works better than dialup. Greg On Mon, 2004-03-01 at 17:44, Jeff McCune wrote: > For one of the project teams I'm involved in, we're traveling to Richmond, > VA for competition for 5 days. In preperation, I decided to solder the USB > cable that connects my Sanyo SCP-4900 cell phone to my Linux notebook in > order to have the 5v USB line charge the phone as well. It was pretty > painless to do, though I would recommend people just buy cables that already > do this, as they're only $15 online now. > > http://www.natecarlson.com/linux/sanyo-4900.php > > I decided to start clean with debian on my notebook to minimize any > ill-effects that might come up at competition, so I did a debootstrap from > Knoppix. Much to my surpise, knoppix already has a complete configuration > for sprinpcs included on the CD. All I had to do was plug the phone into > the USB port, link /dev/modem to /dev/ttyACM0 and then execute pon sprint. > That's pretty sweet. > > With the charging cable, I can now stay online with my notebook for the > entire 8 hour bus ride, as well as the entire day in the pit area in the > arena, so I'm pretty happy. I tested this by leaving the Connection up all > of last night. Worked fine, as long as I stayed sshed somewhere with the > KeepAliveProtocols option turned on. Otherwise, it seems to automatically > shut off after 30 minutes. > > Anyway, if you have SprintPCS and travel a lot, it might be worth your while > to investigate which phones are detected as standard USB acm modems under > Linux. It seems that if you carry the $15 cable and a knoppix CD, you can > easily get on the internet anywhere you have PCS signal. > > It'll be interesting for the 8 hour bus ride and the hotel, since everyone > else has powerbooks... They'll "automatically" get a DHCP address from my > prism2 card acting as an access point and will be routed through the cell > phone. > > Wireless is pretty sweet. Oh, and if it wasn't clear, if you have a Vision > plan with sprint, ppp internet access from a notebook works just fine, and > is completely free (It's considered Vision usage), but it's explicitely > forbidden in the contract you sign. However, I've talked to 3 different > sprint representatives (One in Akron, one in Columbus, one on the phone) and > they all state that Sprint knows people are doing it, and doesn't mind as > long as you don't use it heavily. One of them said that it was a marketing > decision to allow this fact to spread on the net without sprint directly > marketing it as unmetered cellular internet for notebooks. Regardless, it's > not an "official" feature, so if you're thinking of doing this for business, > or are going to rely on it being there, it may go away in the future. > > I've been using it around the holdiays when traveling since last year. It's > great. Around columbus, I get 10-15K/sec (150kbit), and up in north east > ohio I get a solid 200kbit with a steady connection from Akron to Ashtabula. > > Getting closer and closer to my dream of high speed palmtop connectivity... > > Regards, From chris at chrisbaker.net Tue Mar 2 04:12:43 2004 From: chris at chrisbaker.net (Christopher Baker) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Linux firewall In-Reply-To: <20040301211636.GD11576@math.ohio-state.edu> References: <5DBEEED0-6B9D-11D8-95CE-00039317745E@functionalmedia.com> <20040301211636.GD11576@math.ohio-state.edu> Message-ID: <20040302041109.3A93.CHRIS@chrisbaker.net> > If you have the time, I strongly encourage people to roll their own > netfilter iptables scripts to setup their firewall. It was the single most > educational experience in low-level networking I've received. I like the Mandrake Multi Network Firewall. It's working well for us. http://www.mandrakesoft.com/products/mnf ============================================================ FRANK BURNS: Unless we each conform, unless we obey orders, unless we follow our leaders blindly, there is no possible way we can remain free. _M*A*S*H_, "The Novocaine Mutiny" ============================================================ Chris Baker -- www.chrisbaker.net chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com "When you stop growing, you start dying." From chris at chrisbaker.net Tue Mar 2 05:03:30 2004 From: chris at chrisbaker.net (Christopher Baker) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <200403011702.36512.rfunk@funknet.net> References: <40436526.27472.1C045DF@localhost> <200403011702.36512.rfunk@funknet.net> Message-ID: <20040302045812.970B.CHRIS@chrisbaker.net> > Postfix and Qmail are both more secure than Exim, which in turn is more > secure than sendmail. If you're connecting to the internet, security > needs to be a criterion. Most certainly. > Why are you building from source instead of "apt-get install exim" (or > exim-tls)? All the graylisting instructions I have read said to do that. The Debian that I have doesn't want to do Exim 4 either. It likes Exim 3. How can I install Exim 4 with apt-get? > You *really* don't want to leave town right after setting up something like > this. At least, not if you don't want a lynch mob waiting for you when > you come back. That's a good point. If I could leave town with it being ready to go, that would be cool. Then I could do it on the Monday when I get back. Right now, our mail is hosted out side. I want to bring it in. The sooner it gets brought in, the sooner we start saving money. > But not that many people are doing it. Thus it's still experimental. > Spamassassin is not experimental. Graylisting is. More people are using Spam Assassin, and the ones that do seem to love it. But I have always been skeptical of anything that claims to be able to "read" e-mail. I like graylisting because it seems to go more to the root of the problem. It will actually increase the cost of spamming, which is the only real way to cut down on spam. Chris ============================================================ FRANK BURNS: Unless we each conform, unless we obey orders, unless we follow our leaders blindly, there is no possible way we can remain free. _M*A*S*H_, "The Novocaine Mutiny" ============================================================ Chris Baker -- www.chrisbaker.net chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com "When you stop growing, you start dying." From chris at chrisbaker.net Tue Mar 2 07:58:52 2004 From: chris at chrisbaker.net (Christopher Baker) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040301215017.GF11576@math.ohio-state.edu> References: <40436526.27472.1C045DF@localhost> <20040301215017.GF11576@math.ohio-state.edu> Message-ID: <20040302075445.E73D.CHRIS@chrisbaker.net> > In the entire setup, you should be able to use "normal" Debian packages. I > think the only software that I use which wasn't packaged for debian was > MailScanner, and that installs nicely into it's own little section of the > file system. I don't care at all what distribution I use. I just want something that is going to work. I want the project done. > I would start with a "decent" implementation rather than trying to go all > out on the first try. Once you get spamassassin and virus scanning working, > I'd go and then try to integrate other features into the system. Maybe I can use that as a temporary measure. I don't know anymore. I am going to make some phone calls today because I am tired of temporary solutions. Over the last few weeks, I have posted quite a few questions in the Exim and graylisting forums about a problem here and there. I have followed all the instructions and still have problems. I am no longer interested in experimenting. I just want the project done. I have no idea how close I am or how far I am from getting it done. I have no idea how many more problems I am going to encounter. There have already been too many. Chris ============================================================ FRANK BURNS: Unless we each conform, unless we obey orders, unless we follow our leaders blindly, there is no possible way we can remain free. _M*A*S*H_, "The Novocaine Mutiny" ============================================================ Chris Baker -- www.chrisbaker.net chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com "When you stop growing, you start dying." From skippy at skippy.net Tue Mar 2 08:04:13 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040302045812.970B.CHRIS@chrisbaker.net> References: <40436526.27472.1C045DF@localhost><200403011702.36512.rfunk@funknet.net> <20040302045812.970B.CHRIS@chrisbaker.net> Message-ID: <3690.216.136.35.122.1078232653.squirrel@www.skippy.net> Christopher Baker said: > All the graylisting instructions I have read said to do that. The Debian > that I have doesn't want to do Exim 4 either. It likes Exim 3. How can I > install Exim 4 with apt-get? Exim4 is in testing and unstable. Someone said they were using a backported version. Given the choice of using a third-party APT repository or stepping up into "pure" Debian testing or unstable, I would choose testing. Not only have I found testing to be rock-solid; but if you _do_ discover a bug then you can contribute to the value of Debian by reporting the bug and helping the community. >> You *really* don't want to leave town right after setting up something >> like >> this. At least, not if you don't want a lynch mob waiting for you when >> you come back. > > That's a good point. If I could leave town with it being ready to go, > that would be cool. Then I could do it on the Monday when I get back. > Right now, our mail is hosted out side. I want to bring it in. The > sooner it gets brought in, the sooner we start saving money. Saving money won't do any good when email stops working, and you're on vacation! A self-imposed deadline can be dangerous. In your enthusiasm to beat the deadline, you may gloss over some important stuff. A wise man once told me "Measure twice, cut once." Take the time that you've got before vacation to do some more research. Look again at qmail and postfix. Make sure you've chosen the tool you're most comfortable with using and supporting. Since your email is hosted off-site right now, your first goal should be to get email on-site smoothly. After that you can introduce new features. The more knobs you turn, the harder it is to determine which did what. > More people are using Spam Assassin, and the ones that do seem to love > it. But I have always been skeptical of anything that claims to be able > to "read" e-mail. SA doesn't "read" a message for comprehension. It scans a message and looks for certain characteristics: mostly HTML, malformed HTML, the words "opt out", mostly uppercase text, bogus headers, etc. It assigns a "weight" for each of these charactertistics. Some characteristics apply negative weights (ie: makes it less spammy). If the sum of all the weights is higher than your chosen "spam trheshhold", the message is flagged as spam. Whitelisting is accomplished by giving whitelisted addresses a weight of -100, thus ensuring those addresses will _never_ be flagged as spam. SA can also get into Bayesian statistics. I did this for a bit, feeding SA my spam and non-spam sample mboxes. Unfortunately, this made SA flag _way_ too much legitimate mail as spam. I'm not using Bayesian filtering, and SA consistently flags over 60% of my junk mail. > I like graylisting because it seems to go more to the root of the > problem. It will actually increase the cost of spamming, which is the > only real way to cut down on spam. The phrase "security in depth" is appropriate here. Greylisting should be one tool in your arsenal. It's only a matter of time before some well-to-do spammer opens up his own ISP just for spammers. Then they can resend all they want, completely bypassing greylists. MTA content-filtering, procmail filtering, and other tools all add up to create a comprehensive security buffer. It won't be foolproof, and of course "security is inversely proportional to convenience". But a combination of protections will be better than relying on a single one. From skippy at skippy.net Tue Mar 2 08:39:22 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <3690.216.136.35.122.1078232653.squirrel@www.skippy.net> References: <40436526.27472.1C045DF@localhost><200403011702.36512.rfunk@funknet.net> <20040302045812.970B.CHRIS@chrisbaker.net> <3690.216.136.35.122.1078232653.squirrel@www.skippy.net> Message-ID: <3792.216.136.35.122.1078234762.squirrel@www.skippy.net> Scott Merrill said: > Christopher Baker said: >> All the graylisting instructions I have read said to do that. The Debian >> that I have doesn't want to do Exim 4 either. It likes Exim 3. How can I >> install Exim 4 with apt-get? > > Exim4 is in testing and unstable. Someone said they were using a > backported version. Given the choice of using a third-party APT > repository or stepping up into "pure" Debian testing or unstable, I would > choose testing. And for those who don't know, you can find which branch the Debian packages live in at this page: http://www.debian.org/distrib/packages You can also search for which packages own which files. Very handy. From jep200305 at columbus.rr.com Tue Mar 2 08:40:40 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040302042250.GC22453@math.ohio-state.edu> References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> <20040302042250.GC22453@math.ohio-state.edu> Message-ID: <20040302084040.67e2923a.jep200305@columbus.rr.com> Jeff McCune wrote: > I'm not sure if anyone else has mentioned this, > but dynamic DNS might be a good solution. The issue is for helping non-technical users. Anything that requires administration or configuration of any servers, such as but not limited to DHCP, DNS or WINS servers, is beyond the ability and tolerance of non-technical users. For a network which consists of some Win 98SE boxes and a cheap commodity firewall/NAT box, you can just drop in another Windows box without having to mess with anything on the firewall/NAT box or on the other Windows boxes. All the Windows boxes can ping each other by name without any further ado. A Linux box must drop in just as easily. > Not really a solution on a non-dhcp network, Non-technical users can not set up a non-DHCP network, so name resolution for a non-DHCP network is moot and not an issue. From jonadab at bright.net Tue Mar 2 10:23:10 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Message-ID: Jim P writes: > The serious part though, is my consideration of ordinary > non-technical users. For their sake, the solution will have to > something incorporated in a distro whenever Samba is being used, or > maybe even when Samba is not being used. It must require no effort > on their part. Waitasec, since when do ordinary non-technical users name their computers and ping them for network diagnostics? All the ordinary non-technical users that I know carry things from one computer to another on diskettes or CDs, even if the other computer is in the same room. They think "ping" has to do with table tennis or submarines. Still, it would be convenient for powerusers to be able to ping other hosts on the LAN by name without taking any extra steps. I've been using a hosts file, but periodically I discover that the hosts file on one computer or another is out of date and so doesn't include some of the newer systems. > Rob's technique of adding to the resolver library sounds the most > attractive to me, although it's way over my head to either judge > or attempt. I would think that if you wanted it to just work with all applications (ping, traceroute, typing a URI into a web browser, whatever) the most robust thing would be to integrate it with the mechanism that does the name resolution for all those cases, yes. > Perhaps the Samba folks are the ones to provoke an itch among, > which scratched would propagate to all the distros. Or perhaps > provoking an itch amongst someone of a distro who would push some > patches toward Samba and pull the result back for seamless > integration into a distro. > > Who else should this itch be propagated to? Who maintains the resolver? What project is that a part of? Is that part of the kernel? Or is it part of the standard C library? I don't know exactly, but I would think those would be the people to talk to about this matter. The Samba people would have to coordinate with them, I would think. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Tue Mar 2 10:33:22 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] apt-get In-Reply-To: <4043DADD.7030901@skippy.net> References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> <20040301211207.GC11576@math.ohio-state.edu> <4043DADD.7030901@skippy.net> Message-ID: Scott Merrill writes: > See the mount manpage. > > -o Options > remount > "Attempt to remount an already-mounted file system. This is > commonly used to change the mount flags for a file system, > especially to make a readonly file system write- able. It does not > change device or mount point." Yeah, but can you do this while processes are running that hold open files and inodes on that filesystem? I know you can copy and move files out from under running processes, but I was under the impression that was due to the separation between the file directory information and the inodes (a distinction FAT doesn't have). Also, one time when I was going to junk a particular install, just to see how far it would get, I tried to rm -r /usr, and the system froze pretty solid (as I'd guessed it would) well before the operation completed. This was a while ago, on a pretty old system (2.0 series kernel I believe), but I guess I didn't figure something like that would change quickly. Jeff McCune seems to have actually *done* this, so apparently I was wrong -- but I'm interested in knowing where my reasoning failed. Is there a fundamental difference between removing files versus unmounting the filesystem? Or does remounting not actually unmount, just change the flags? Or has this changed in more recent kernels? -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jep200305 at columbus.rr.com Tue Mar 2 11:16:08 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Message-ID: <20040302111608.53063c30.jep200305@columbus.rr.com> jonadab@bright.net wrote: > Jim P writes: > > > The serious part though, is my consideration of ordinary > > non-technical users. For their sake, the solution will have to > > something incorporated in a distro whenever Samba is being used, or > > maybe even when Samba is not being used. It must require no effort > > on their part. > > Waitasec, since when do ordinary non-technical users name their > computers and ping them for network diagnostics? You're right. Non-technical users don't use ping. Ping is the simplest example of a classic UNIX style network command. I didn't want to confuse the issue with more complicated network commands. The name resolution has to be working seamlessly before the non-technical users can even consider trying the UNIX network commands. > All the ordinary > non-technical users that I know carry things from one computer to > another on diskettes or CDs, even if the other computer is in the same > room. I know several non-technical users who have more than one computer hooked up with a firewall/NAT box to high-speed internet access. It's becoming common. There certainly are still plenty of non-technical users who have stand-alone computers. > Still, it would be convenient for powerusers to be able to ping other > hosts on the LAN by name without taking any extra steps. I've been > using a hosts file, but periodically I discover that the hosts file on > one computer or another is out of date and so doesn't include some of > the newer systems. And on a DHCP network (which is what non-technical users have), /etc/hosts is a poor option. From rfunk at funknet.net Tue Mar 2 11:35:48 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <20040301230430.3ec6cb58.jep200305@columbus.rr.com> References: <200403011750.40710.colug@cencula.com> <4043DB51.9060302@skippy.net> <20040301230430.3ec6cb58.jep200305@columbus.rr.com> Message-ID: <200403021135.48228.rfunk@funknet.net> Jim P wrote: > Yet another completely different approach, would be to run a PC > emulator on an ISA Linux box and configure the emulator so that > access to the programmer's I/O ports is passed to actual hardware, > instead of being faked. This might actually work. > One would certainly have to read the emulator manuals to > see how to configure it for actual access to real hardware. > > Check out VMware and Bochs http://bochs.sourceforge.net/ > A virtual machine like Plex86 http://plex86.sourceforge.net/ > might also work, _fast_. I had a similar thought, but I would add dosemu to the list. It gives DOS a BIOS layer that communicates with Linux, which can then be connected to the hardware. It's not quite as heavy as the full-machine emulators, and should be faster. http://www.dosemu.org/ -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From rfunk at funknet.net Tue Mar 2 11:55:16 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Message-ID: <200403021155.16033.rfunk@funknet.net> Jonadab the Unsightly One wrote: > Who maintains the resolver? ?What project is that a part of? ?Is that > part of the kernel? ?Or is it part of the standard C library? ?I don't > know exactly, but I would think those would be the people to talk to > about this matter. ?The Samba people would have to coordinate with > them, I would think. Actually it's not necessary to mess with the resolver library itself. Anyone familiar with the old NIS/YP network name systems has seen how /etc/nsswitch.conf can be used to configure where the system looks for the information normally provided by such files as /etc/passwd, /etc/group, and /etc/hosts. If you look at the winbind page I pointed to, it talks about the ability to link into the name service switch (NSS) by adding a library like "/lib/libnss_mynameservice.so". Then you'd add "mynameservice" to lines in /etc/nsswitch.conf. The samba/winbind people have already made a libnss_winbind.so that handles users and groups. Apparently the remaining work would be to make that library handle hostname resolution too. After that it's just a matter of making it easily installable and/or installed in distros by default. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Tue Mar 2 11:55:34 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <200403021135.48228.rfunk@funknet.net> References: <200403011750.40710.colug@cencula.com> <4043DB51.9060302@skippy.net> <20040301230430.3ec6cb58.jep200305@columbus.rr.com> <200403021135.48228.rfunk@funknet.net> Message-ID: <20040302115534.415b55ed.jep200305@columbus.rr.com> Rob wrote: > Jim P wrote: > > Yet another completely different approach, would be to run a PC > > emulator on an ISA Linux box and configure the emulator so that > > access to the programmer's I/O ports is passed to actual hardware, > > instead of being faked. > > Check out VMware and Bochs > > Plex86 > I had a similar thought, but I would add dosemu to the list. It gives DOS > a BIOS layer that communicates with Linux, which can then be connected to > the hardware. It's not quite as heavy as the full-machine emulators, and > should be faster. > http://www.dosemu.org/ Wow. I'm already using dosemu to run an old DOS compiler from Linux make, so this is attractive for using my device programmer and a flash programming utility directly from Linux without having to transfer the file to a Windows box to burn. From holbert.13 at osu.edu Tue Mar 2 11:57:47 2004 From: holbert.13 at osu.edu (Rick Holbert) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] Remote operation of Free-DOS box, preferably over network In-Reply-To: <200403021135.48228.rfunk@funknet.net> References: <200403011750.40710.colug@cencula.com> <20040301230430.3ec6cb58.jep200305@columbus.rr.com> <200403021135.48228.rfunk@funknet.net> Message-ID: <200403021157.47494.holbert.13@osu.edu> I second Rob's recommendation of dosemu and also recommend a look at dosbox: http://freshmeat.net/projects/dosbox/ On Tuesday 02 March 2004 11:35, Rob Funk wrote: > I had a similar thought, but I would add dosemu to the list. It gives DOS > a BIOS layer that communicates with Linux, which can then be connected to > the hardware. It's not quite as heavy as the full-machine emulators, and > should be faster. > http://www.dosemu.org/ From rfunk at funknet.net Tue Mar 2 12:11:42 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:36 2005 Subject: [COLUG] dosemu In-Reply-To: <20040302115534.415b55ed.jep200305@columbus.rr.com> References: <200403011750.40710.colug@cencula.com> <200403021135.48228.rfunk@funknet.net> <20040302115534.415b55ed.jep200305@columbus.rr.com> Message-ID: <200403021211.42251.rfunk@funknet.net> Jim P wrote: > I'm already using dosemu to run an old DOS compiler from > Linux make This sounds intriguing. I'd be interested in hearing how you set up the automated communication between Linux and DOS. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From whooper at freeshell.org Tue Mar 2 12:28:33 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <200403021155.16033.rfunk@funknet.net> References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> <200403021155.16033.rfunk@funknet.net> Message-ID: <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> Rob Funk said: > The samba/winbind people have already made a libnss_winbind.so that > handles > users and groups. Apparently the remaining work would be to make that > library handle hostname resolution too. Are you sure this isn't already there in Samba > 3? http://us3.samba.org/samba/docs/man/winbindd.8.html "The following simple configuration in the /etc/nsswitch.conf file can be used to initially resolve hostnames from /etc/hosts and then from the WINS server. hosts: files wins" It also mentions being able to resolve via broadcast (not sure if that is part of this set up or not). -- William Hooper From rfunk at funknet.net Tue Mar 2 12:47:10 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> References: <200402292351.58732.rfunk@funknet.net> <200403021155.16033.rfunk@funknet.net> <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> Message-ID: <200403021247.10753.rfunk@funknet.net> William Hooper wrote: > Rob Funk said: > > Apparently the remaining work would be to make that > > library handle hostname resolution too. > > Are you sure this isn't already there in Samba > 3? > > http://us3.samba.org/samba/docs/man/winbindd.8.html > "The following simple configuration in the /etc/nsswitch.conf file can > be used to initially resolve hostnames from /etc/hosts and then from the > WINS server. > hosts: files wins" Hey, cool! That makes it just a matter of installation and configuration! I guess the page I found was for the older version of Samba. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Tue Mar 2 12:54:33 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] dosemu In-Reply-To: <200403021211.42251.rfunk@funknet.net> References: <200403011750.40710.colug@cencula.com> <200403021135.48228.rfunk@funknet.net> <20040302115534.415b55ed.jep200305@columbus.rr.com> <200403021211.42251.rfunk@funknet.net> Message-ID: <20040302125433.59baed64.jep200305@columbus.rr.com> Rob wrote: > Jim P wrote: > > I'm already using dosemu to run an old DOS compiler from > > Linux make > > This sounds intriguing. I'd be interested in hearing how you set up the > automated communication between Linux and DOS. I've been away from that stuff for a few months. I went back to play with that stuff, but now I can not replicate what I did, but remember that it was quite unsophisticated. I was (re)invoking dosemu for each time I needed to run the (DOS) compiler or linker. There were some issues about DOS not setting file timestamps correctly, which confused make. I'd often have to help by deleting a target file. Nuts. Jim From mark.aufdencamp at prospicien.com Tue Mar 2 13:01:32 2004 From: mark.aufdencamp at prospicien.com (Mark Aufdencamp) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] NetBIOS Name Service and DNS Integration Message-ID: I've been way to busy to contribute the last three months and have a day of downtime while awaiting some overnight components. I was going to attend a M$ seminar on .NET security this afternoon, but instead I thought I'd blow it off and contribute my two cents worth the knowledge on NetBIOS, NetBT (NetBIOS over TCP), and NBNS (NetBIOS Name Services). In the beginning there was LAN Manager... Anyone remember the old 3Com 3Servers? This was the first LAN Manager server I was exposed to. LAN Manager existed before OS/2, Windows for Workgroups, or Windows NT. Microsoft and IBM came up with the NetBIOS networking specification in the mid-eighties. It specified an API of DOS software interrupts that provided network communications and network file/record locking. It was designed for a single broadcast domain with no more than 255 workstations. Many low end peer to peer networks utilized it as a core technology. However, this was extremely limiting in the enterprise. Thus NetBEUI (NetBIOS Extended User Interface) and NetBT were birthed. NetBEUI utilizes 802.2 LLC and NBFP (NetBIOS Frame Protocol) to provide "..It performs session establishment, multiplexing, and termination. It also performs message segmentation, delimiting, assembly, and acknowledgment." - Microsoft Windows NT Server Networking Guide When a typical Microsoft network is configured out of the box, NetBEUI is utilized as the default protocol. The default method of name resolution with this default protocol configuration is registration with a NetBIOS Master Browser via 802.2 LLC Broadcast. A workstation within the LAN broadcast domain will always be designated a Master Browser. When one opens Network Neighborhood on a M$ OS, the workstation list from the Master Browser is displayed. The workstation that will receive the honor of becoming the Master Browser is determined by an election. The election process is rigged much like ballot counting in Florida. The latest and greatest M$ operating system will be weighted to win. When one configures SAMBA, an OS identifier is specified to rank the SAMBA server in this heirarchy. This works well for a small home or office network, and requires virtually no configuration by the user other than unique machine names. However, it becomes a loser when applied to the enterprise. Connecting offices over a WAN required that the networks be bridged. This is typically a bad thing and may be downright impossible with some technologies. Enter NetBIOS over TCP/IP. With the advent of OS/2 and later with NT, M$ defined a method for encapsulating NetBIOS in TCP. This was great. Now NetBIOS could be sent over the WAN, and one could see workstations in other offices around the world in network neighborhood. One slight problem though, how to coordinate multiple Master Browsers? (one for each broadcast domain) Thus we arrive at the NetBIOS Name Server. This is more commonly refered to as a WINS server in the M$ world. A NetBIOS name server supports a point-to-point name registration and resolution, much like a Dynamic DNS server, in addition to the broadcast mechanism. It also supports replication with other NBNS's and Master Browsers. The method of name registration and resolution utilized by a M$ workstation is determined by a value in a registry key. I personally utilize a hybrid mode, where point-to-point communications with a NBNS is first utilized, and failing that the workstation falls back to the broadcast mechanism. This method and the IP address's are handed out as a part of DHCP. Thus a DHCP server is responsible for providing the following: IP Address and Subnet, Default Gateway, DNS Server address(s), NBNS registration/resolution method, NBNS Server address(s). Beginning with NT Server, the M$ DNS server and WINS server can be integrated to provide a common name resolution mechanism. Essentially, the DNS server will query the WINS server for NetBT workstation names. This works quite well for private name services. However, having seen a few M$ DNS server exploits perpetrated on my network, I would not run M$ DNS as a public name server. SAMBA contains a NBNS as a component. It functions much like WINS. It could conceivable be enhanced to proxy NetBIOS name registrations and resolution request through to a dynamic DNS server. Thus providing Name to IP mappings for non NetBIOS aware host. The reverse capability already exist in the M$ NetBIOS clients. They can be configured to utilize both methods, either the NBNS or DNS server. This brings us to a well known problem of split network horizons. When multiple M$ workstations utilize different NBNS registration/resolution methods strange things can happen. I've personally seen and corrected environments where W95 workstations were utilizing NetBEUI and NT workstations were using a WINS server and they couldn't see each other. Properly configuring NetBIOS Name Services can lead to really great things in the enterprise. Like a thousand workstations spreads over the world in network neighborhood all nicely segregated and browse able. As an additional side note, it may behoove one to utilize different NetBIOS vs DNS Host names. It becomes much easier to troubleshoot which name service and transport mechanism is being utilized when one can ping over different transports by using varying names. For anyone interested in a little more fun, play around with "nbtstat" on your M$ workstation. The -a and -A are particularly interesting. Yes, NetBT does utilize specific TCP ports that must be blocked on the firewall or bad things will happen. Hope I've enlightened some. Mark Aufdencamp, KoolBean Aufdencamp Information Services, Inc. Mark@Aufdencamp.com From jep200305 at columbus.rr.com Tue Mar 2 12:59:49 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] Integrating Linux with crudest Windows networks In-Reply-To: <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> References: <200402292351.58732.rfunk@funknet.net> <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> <200403021155.16033.rfunk@funknet.net> <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> Message-ID: <20040302125949.7b6cef40.jep200305@columbus.rr.com> William Hooper wrote: > Rob Funk said: > > > The samba/winbind people have already made a libnss_winbind.so that > > handles > > users and groups. Apparently the remaining work would be to make that > > library handle hostname resolution too. > > Are you sure this isn't already there in Samba > 3? > > http://us3.samba.org/samba/docs/man/winbindd.8.html > "The following simple configuration in the /etc/nsswitch.conf file can be > used to initially resolve hostnames from /etc/hosts and then from the WINS > server. As Rob wrote, that takes care of what he was looking for. Do you see anything that takes care of folks without WINS servers? From tom at functionalmedia.com Tue Mar 2 13:03:07 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] Integrating Linux with Windows networks In-Reply-To: <20040301215519.1fcc1e8a.jep200305@columbus.rr.com> Message-ID: Why not do it the microsoft way ?? Have the linux box detect that it is in a MS environment. It then installs samba configures itself to win all elections sets itself up as master host or whatever windows calls this. The linux box also perhaps as the WINS server. It then uses an email or any one of a multitude of exploit options, to infect all the other machines with a virus which opens up a backdoor to all the machines on the network. Utilizing this backdoor the linux machine can chose to either tolerate the windows machines on the network or set up a netboot install of linux on all the machines on the network. Each machine on the home network proceeds to install linux and VIOLA a uniform environment is achieved. I think that is how Bill G would solve this problem, that is if he worked for us. Perhaps we must think like the enemy in order to win this war. This is how windows treats an ext3 file system on my little usb drive, so why should we treat rogue protocols with any more respect than they deserve? By the way I am just kidding..Jim P has proposed an interesting and challenging issue that deserves attention. Tom On Monday, March 1, 2004, at 09:55 PM, Jim P wrote: > Thanks Rob for remembering this issue and continuing to think about it. > Thanks to Scott, Joe and Jeff for trying to help. > > For myself, I just cope by using nmblookup, and Jeff's hack helps. > > The serious part though, is my consideration of ordinary non-technical > users. For their sake, the solution will have to something > incorporated > in a distro whenever Samba is being used, or maybe even when Samba is > not > being used. It must require no effort on their part. > > Rob's technique of adding to the resolver library sounds the most > attractive to me, although it's way over my head to either judge > or attempt. > > Perhaps the Samba folks are the ones to provoke an itch among, > which scratched would propagate to all the distros. Or perhaps > provoking an itch amongst someone of a distro who would push some > patches toward Samba and pull the result back for seamless > integration into a distro. > > Who else should this itch be propagated to? > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From skippy at skippy.net Tue Mar 2 13:17:33 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] Integrating Linux with crudest Windows networks In-Reply-To: <20040302125949.7b6cef40.jep200305@columbus.rr.com> References: <200402292351.58732.rfunk@funknet.net><20040301215519.1fcc1e8a.jep200305@columbus.rr.com><200403021155.16033.rfunk@funknet.net><3147.209.239.137.161.1078248513.squirrel@209.239.137.161> <20040302125949.7b6cef40.jep200305@columbus.rr.com> Message-ID: <1103.216.136.35.122.1078251453.squirrel@www.skippy.net> Jim P said: > William Hooper wrote: > >> Rob Funk said: >> >> > The samba/winbind people have already made a libnss_winbind.so that >> > handles >> > users and groups. Apparently the remaining work would be to make that >> > library handle hostname resolution too. >> >> Are you sure this isn't already there in Samba > 3? >> >> http://us3.samba.org/samba/docs/man/winbindd.8.html >> "The following simple configuration in the /etc/nsswitch.conf file can >> be >> used to initially resolve hostnames from /etc/hosts and then from the >> WINS >> server. > > As Rob wrote, that takes care of what he was looking for. > > Do you see anything that takes care of folks without WINS servers? winbind has either been better documented, or features have been added since last I looked: http://us2.samba.org/samba/docs/man/winbindd.8.html Note the caveat "nmbd(8) must be running on the local machine for winbindd to work". You'll need to run Samba on your GNU/Linux system, and configure it to use winbind, which should in turn interrogate the local broadcast domain for NetBIOS name resolution. See also: http://us2.samba.org/samba/docs/man/wbinfo.1.html From mccune at math.ohio-state.edu Tue Mar 2 13:19:48 2004 From: mccune at math.ohio-state.edu (Jeff McCune) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] apt-get In-Reply-To: References: <40436DDA.3000205@bugs.osu.edu> <69E41634-6BA8-11D8-95CE-00039317745E@functionalmedia.com> <20040301211207.GC11576@math.ohio-state.edu> <4043DADD.7030901@skippy.net> Message-ID: <20040302181948.GA22499@math.ohio-state.edu> On Tue, Mar 02, 2004 at 10:33:22AM -0500, Jonadab the Unsightly One enlightened us: > Scott Merrill writes: > > > See the mount manpage. > > > > -o Options > > remount > > "Attempt to remount an already-mounted file system. This is > > commonly used to change the mount flags for a file system, > > especially to make a readonly file system write- able. It does not > > change device or mount point." > > Yeah, but can you do this while processes are running that hold open > files and inodes on that filesystem? Yes, you certainly can. That's the whole point. > I know you can copy and move files out from under running processes, > but I was under the impression that was due to the separation between > the file directory information and the inodes (a distinction FAT > doesn't have). You can do more than that. You can even remove files held open by running processes. They'll happily run until completion. > Also, one time when I was going to junk a particular install, just to > see how far it would get, I tried to rm -r /usr, and the system froze > pretty solid (as I'd guessed it would) well before the operation > completed. This was a while ago, on a pretty old system (2.0 series > kernel I believe), but I guess I didn't figure something like that > would change quickly. I don't understand what rm -rf /usr causing a system hang would tell you about mount -o rw,remount /usr. In the former action, you're removing (possibly) hundreds of libraries and binaries that running software expects to have access to. In the other situation you're simply locking processes from further writes to the files they have open. One is considerably more dangerous than the other. Also, remember that there should be very few, if any, files in /usr open for writing. > Jeff McCune seems to have actually *done* this, so apparently I was > wrong -- but I'm interested in knowing where my reasoning failed. > Is there a fundamental difference between removing files versus > unmounting the filesystem? Or does remounting not actually unmount, > just change the flags? Or has this changed in more recent kernels? > Yes, there *is* a fundamental difference between removing files and remounting a file system. Again, I don't see how you're relating rm -rf /usr/ to a simple remount. It'd be like comparing a swift blow to the head with a shovel to asking kindly for your keys back. They yield completely different results. I originally got this trick from Tridge, who's one of the samba developers. He mentioned in his Tivo hacking docs how you need to mount -o remount,rw / in order to do stuff on the tivo, since it's root FS is mounted ro on boot. That, and the fact that there's no blatant warnings about it in the man page leads me to believe it operates as documented. Regards, -- Jeff McCune Department of Mathematics The Ohio State University (614) 292-4962 gpg --keyserver pgp.mit.edu --recv-key BAF3211A -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 185 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040302/a21ff9ec/attachment.bin From brian at bluedogs-online.com Tue Mar 2 13:48:34 2004 From: brian at bluedogs-online.com (Brian Morgan) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS Message-ID: <4044D702.7020605@bluedogs-online.com> http://www.catb.org/~esr/writings/cups-horror.html Hello all, I am new to the area and to this list. Just wanted to say hi and to pass on the rant by ESR, I found it very interesting because it echos what my friends and family feel about using linux. Why switch when windows is just so easy. Hope you enjoy the article, Brian Morgan CompE From blata at extent0006.entomology.ohio-state.edu Tue Mar 2 14:21:38 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <4044D702.7020605@bluedogs-online.com> References: <4044D702.7020605@bluedogs-online.com> Message-ID: <4044DEC2.5040602@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 wow I think i might now know whats wrong with my home setup. Brian Morgan spewed verbiage on 03/02/2004 01:48 PM: | http://www.catb.org/~esr/writings/cups-horror.html | | Hello all, I am new to the area and to this list. | | Just wanted to say hi and to pass on the rant by ESR, I found it very | interesting because it echos what my friends and family feel about using | linux. Why switch when windows is just so easy. | | Hope you enjoy the article, | | Brian Morgan | CompE | | | _______________________________________________ | colug mailing list | colug@colug.net | http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFARN7Cv+6+qSFv3TURAg+JAJ9qNtPjEaJ1Vln6FjtTHhbm20em2gCeJEmf dkUXxsJ8KUm0dQ/DNz1mC2A= =OToB -----END PGP SIGNATURE----- From colug at cencula.com Tue Mar 2 19:35:31 2004 From: colug at cencula.com (Michael D. Cencula) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <4044D702.7020605@bluedogs-online.com> References: <4044D702.7020605@bluedogs-online.com> Message-ID: <200403021935.31139.colug@cencula.com> On Tuesday 02 March 2004 1:48 pm, Brian Morgan wrote: > http://www.catb.org/~esr/writings/cups-horror.html > > Hello all, I am new to the area and to this list. > > Just wanted to say hi and to pass on the rant by ESR, I found it very > interesting because it echos what my friends and family feel about using > linux. Why switch when windows is just so easy. Now, there's a troll if I've ever heard one. > Hope you enjoy the article, > > Brian Morgan > CompE > > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug From brucehohl at yahoo.com Tue Mar 2 20:46:55 2004 From: brucehohl at yahoo.com (Bruce Hohl) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <4044D702.7020605@bluedogs-online.com> Message-ID: <20040303014655.69383.qmail@web13121.mail.yahoo.com> --- Brian Morgan wrote: > http://www.catb.org/~esr/writings/cups-horror.html > Hope you enjoy the article, > A little over a year ago I set up my first linux box and had printing problems. I put a post on www.linuxprinting.org and none other than Till Kamppeter returned my posts until I got my printer working. So a complete newbie (me) got help directly from one of the CUPS authors and got the printer going likely in a fraction of the time it took to write the above rant. I think that author (Eric Raymond) should consider all the good CUPS has brought to the open source world - and offer to write a cups tutorial instead of a rant. __________________________________ Do you Yahoo!? Yahoo! Search - Find what you’re looking for faster http://search.yahoo.com From tnoe at mailsnare.net Tue Mar 2 21:31:00 2004 From: tnoe at mailsnare.net (Thomas J. Noe) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <20040303014655.69383.qmail@web13121.mail.yahoo.com> References: <20040303014655.69383.qmail@web13121.mail.yahoo.com> Message-ID: <200403022131.03443.tnoe@mailsnare.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Except that I think that the point of ESR's essay was both amusement for those of us who have dealt with the same demon (CUPS specifically), and also to inspire open source authors (CUPS and everyone else) to try and put as much thought and consideration into the installation and configuration process of their software as they did when writing the software itself. You can have the best piece of software in the world, but if no one can install and configure it properly, then it isn't going to do anyone any good. Just my $0.02. BTW, that's cool that you actually got help from one of the project's authors. On Tuesday 02 March 2004 20:46, Bruce Hohl wrote: > --- Brian Morgan wrote: > > http://www.catb.org/~esr/writings/cups-horror.html > > Hope you enjoy the article, > > A little over a year ago I set up my first linux box > and had printing problems. I put a post on > www.linuxprinting.org and none other than Till > Kamppeter returned my posts until I got my printer > working. > > So a complete newbie (me) got help directly from one > of the CUPS authors and got the printer going likely > in a fraction of the time it took to write the above > rant. I think that author (Eric Raymond) should > consider all the good CUPS has brought to the open > source world - and offer to write a cups tutorial > instead of a rant. > > __________________________________ > Do you Yahoo!? > Yahoo! Search - Find what you?re looking for faster > http://search.yahoo.com > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug - -- Best wishes, Tom E: tnoe@mailsnare.net P: tnoe@vtext.com GPG keyID 0x9ACE08E1 gpg --keyserver pgp.mit.edu --recv-keys 9ACE08E1 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFARUNnNFfoH5OP+5oRAhvDAJ4jUV2Hhujg2hMUfCj2JmMEQAcyGQCbBI0z 1mYjJlUj/jH8tMrMPXzjzLQ= =+E5C -----END PGP SIGNATURE----- From brian at bluedogs-online.com Tue Mar 2 22:15:41 2004 From: brian at bluedogs-online.com (Brian Morgan) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <200403021935.31139.colug@cencula.com> References: <4044D702.7020605@bluedogs-online.com> <200403021935.31139.colug@cencula.com> Message-ID: <40454DDD.7080509@bluedogs-online.com> Good evening, Michael D. Cencula wrote: >On Tuesday 02 March 2004 1:48 pm, Brian Morgan wrote: > > >>http://www.catb.org/~esr/writings/cups-horror.html >> >>Hello all, I am new to the area and to this list. >> >>Just wanted to say hi and to pass on the rant by ESR, I found it very >>interesting because it echos what my friends and family feel about using >>linux. Why switch when windows is just so easy. >> >> > >Now, there's a troll if I've ever heard one. > > > First did you read the article? Second that is not my personal feelings but it is what I hear day in and day out in the corporate trenches. Not every statement made supporting a OS other then Linux, or a statement made highlighting a flaw in the software is meant as an attack on your way of life. I apologize to the list if my statement appeared as trolling flamebait, I did not intentionally make my first post to the list a troll. I was hoping to get some local discussion going on what people would like to see changed in linux. Any suggestions? I was more going for point that as new features are added into linux we can't forget who we want to be able to use the software, everyone. As I replied to a fellow colugger off the list, can you imagine your mother trying to print off the latest food.com recipe with a OoTB SuSE/Debian//NetBSD install? No. Lets do a quick survey of coluggers here and see what we can find: 1. What is the most common task that your friends and families do on their computer, with a microsoft OS installed? a. Print off web pages, emails, and text documents to a plug and play printer. 2. How long did it take to setup? a. About as long as it took the computer to start. 3. Is this functionality be duplicated in linux? a. yes 4. How long did it take for you to setup? a. I took me a weekend to get linux workstation and freebsd workstation talking to a HP Laserjet installed on a XP box. Anyways I was just hoping to see if anyone else out there has spent an evening cursing at their sendmail.cf/snort.conf/httpd.conf file, or if it is just me. I am curious to see what the response is. > > Brian Morgan CompE If linux is going to be for everyone what are you doing to help get it there? From jep200305 at columbus.rr.com Tue Mar 2 22:46:41 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040303014655.69383.qmail@web13121.mail.yahoo.com> References: <4044D702.7020605@bluedogs-online.com> <20040303014655.69383.qmail@web13121.mail.yahoo.com> Message-ID: <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> Bruce Hohl wrote: > --- Brian Morgan wrote: > > http://www.catb.org/~esr/writings/cups-horror.html > A little over a year ago I set up my first linux box > and had printing problems. I put a post on > www.linuxprinting.org and none other than Till > Kamppeter returned my posts until I got my printer > working. > > So a complete newbie (me) got help directly from one > of the CUPS authors and got the printer going likely > in a fraction of the time it took to write the above > rant. I think that author (Eric Raymond) should > consider all the good CUPS has brought to the open > source world - and offer to write a cups tutorial > instead of a rant. Writing a tutorial would not have helped the non-technical users. For a non-technical user, having to consult a tutorial is a failure of the software. Instead software needs to work so that a non-technical user does not need a tutorial. ESR is trying to change the way all OSS software is done, not just CUPS. The problem is that the people who have the skills to make the programs, don't understand that their software is difficult for non-technical people to use. From jonadab at bright.net Tue Mar 2 22:53:21 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:37 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <200403021935.31139.colug@cencula.com> References: <4044D702.7020605@bluedogs-online.com> <200403021935.31139.colug@cencula.com> Message-ID: <1xoak2mm.fsf@jonadab.homeip.net> "Michael D. Cencula" writes: > On Tuesday 02 March 2004 1:48 pm, Brian Morgan wrote: > > http://www.catb.org/~esr/writings/cups-horror.html > > > > Hello all, I am new to the area and to this list. > > > > Just wanted to say hi and to pass on the rant by ESR, I found it very > > interesting because it echos what my friends and family feel about using > > linux. Why switch when windows is just so easy. > > Now, there's a troll if I've ever heard one. Certain things about Windows are easy, yes, easier than on other systems. Not enough things or the right things to get me to use Windows as my primary system, mind you, but some things are easier. What's much much harder is customization, and that happens to be important to me. But yeah, ESR has a good point. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Tue Mar 2 22:56:37 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <20040303014655.69383.qmail@web13121.mail.yahoo.com> References: <20040303014655.69383.qmail@web13121.mail.yahoo.com> Message-ID: Bruce Hohl writes: > So a complete newbie (me) got help directly from one of the CUPS > authors and got the printer going likely in a fraction of the time > it took to write the above rant. That's one of the good things about the OSS . > I think that author (Eric Raymond) should consider all the good CUPS > has brought to the open source world - and offer to write a cups > tutorial instead of a rant. You missed the point of the article. CUPS was his *example*, but he was speaking generally. Writing a CUPS tutorial or, indeed, a better CUPS configuration GUI, would not solve the larger problem. The rant probably won't solve it either, but it might help a little, and in any case it's a nice gesture. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From blata at extent0006.entomology.ohio-state.edu Tue Mar 2 23:25:32 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: References: <20040303014655.69383.qmail@web13121.mail.yahoo.com> Message-ID: <40455E3C.1050601@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am I mistaken or is the configuration software not a cups piece of software. Isn't it a RedHat piece of soft ware? redhat-configure-printers if I'm not mistaken. I do agree with the difficulty of setting up cups especaly for sharing with samba client (xp, 9x, etc.) by the way I finally got it working today thanks to the ranting of the artical in question. Sorry beer goggles. I forgot my point. You wrote this to me on 03/02/2004 10:56 PM: | Bruce Hohl writes: | |> So a complete newbie (me) got help directly from one of the CUPS |> authors and got the printer going likely in a fraction of the |> time it took to write the above rant. | | | That's one of the good things about the OSS . | |> I think that author (Eric Raymond) should consider all the good |> CUPS has brought to the open source world - and offer to write a |> cups tutorial instead of a rant. | | | You missed the point of the article. CUPS was his *example*, but | he was speaking generally. Writing a CUPS tutorial or, indeed, a | better CUPS configuration GUI, would not solve the larger problem. | The rant probably won't solve it either, but it might help a | little, and in any case it's a nice gesture. | - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFARV47v+6+qSFv3TURAmHfAJ99GBj+OV3Bs6OuUnT5rTygPJ670wCePHbt WKUfAy3PrhjzVoUUUuzpEPU= =3nnW -----END PGP SIGNATURE----- From chris at chrisbaker.net Wed Mar 3 00:09:29 2004 From: chris at chrisbaker.net (Christopher Baker) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Debian/Exim/Perl/mysql expert needed In-Reply-To: <20040301214342.GE11576@math.ohio-state.edu> References: <404355AE.12999.183D97D@localhost> <20040301214342.GE11576@math.ohio-state.edu> Message-ID: <20040303000833.BB72.CHRIS@chrisbaker.net> > I haven't heard about greylisting. Any links? Here is one: http://projects.puremagic.com/greylisting/ Chris ============================================================ FRANK BURNS: Unless we each conform, unless we obey orders, unless we follow our leaders blindly, there is no possible way we can remain free. _M*A*S*H_, "The Novocaine Mutiny" ============================================================ Chris Baker -- www.chrisbaker.net chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com "When you stop growing, you start dying." From rfunk at funknet.net Wed Mar 3 02:47:41 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <40454DDD.7080509@bluedogs-online.com> References: <4044D702.7020605@bluedogs-online.com> <200403021935.31139.colug@cencula.com> <40454DDD.7080509@bluedogs-online.com> Message-ID: <200403030247.41783.rfunk@funknet.net> Brian Morgan wrote: > Michael D. Cencula wrote: > >On Tuesday 02 March 2004 1:48 pm, Brian Morgan wrote: > >>Just wanted to say hi and to pass on the rant by ESR, I found it very > >>interesting because it echos what my friends and family feel about > >> using linux. Why switch when windows is just so easy. > > > >Now, there's a troll if I've ever heard one. > > First did you read the article? I did. And I largely agreed with it. (You can find my comments, under the name rfunk, attached to lwn.net's link to the article: http://lwn.net/Articles/73178/) Yet when I read your Windows line I had a single thought: Do not feed the troll. Sorry, but that's how your message came across. > Not every statement made supporting a OS other then Linux, or a > statement made highlighting a flaw in the software is meant as an attack > on your way of life. True. Constructive criticism is useful, which is why ESR's rant is valuable. > I apologize to the list if my statement appeared as trolling flamebait, > I did not intentionally make my first post to the list a troll. I was > hoping to get some local discussion going on what people would like to > see changed in linux. Any suggestions? ESR has credibility on the issue because he is a longtime Linux/Unix developer who has taken a leadership role in the community. So when he points out a failing in that community which is his community too, it has more weight than it would coming from some random users. > As I replied to a fellow colugger off the list, can you imagine your > mother trying to print off the latest food.com recipe with a OoTB > SuSE/Debian//NetBSD install? No. Actually I can. Local printing generally isn't a problem. Network printing is where things get tricky. > 1. What is the most common task that your friends and families do on > their computer, with a microsoft OS installed? web browsing and reading/writing email. > a. Print off web pages, emails, and text documents to a plug and > play printer. No, in fact. > 4. How long did it take for you to setup? > > a. I took me a weekend to get linux workstation and freebsd > workstation talking to a HP Laserjet installed on a XP box. Not a fair comparison. Connect that Laserjet directly to the Linux box and it'll be much easier to get set up, and in turn a bit harder to tell Windows to print to it. Local printing is easy. Network printing is hard. Some systems make it harder than others. (CUPS can be easy if you're lucky, or hard if you're unlucky.) > Anyways I was just hoping to see if anyone else out there has spent an > evening cursing at their sendmail.cf/snort.conf/httpd.conf file, or if > it is just me. This doesn't fit with your "Linux doesn't work for normal users" theme. How many normal users mess with snort at all? Hardly any. Sendmail? I think you'll find just about anyone here recommending any of at least three alternatives as easier to set up, and most users don't use any of the four since they just tell their mail client to talk directly to their ISP's mail server. I can actually write and understand sendmail.cf rules reasonably well, yet I generally use and recommend other options. Apache? Most normal people pay other people (techies who like messing with httpd.conf) to run their web servers, and just worry about content. > I am curious to see what the response is. Oops, I fed the troll. > If linux is going to be for everyone what are you doing to help get it > there? I don't see it as my job to evangelize Linux, nor to keep Linux confined to a select group of people. I just keep "scratching my own itch" and attacking problems that interest me, and maybe that'll help someone else down the line. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From peter.king at utoronto.ca Wed Mar 3 10:24:06 2004 From: peter.king at utoronto.ca (Peter King) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> References: <4044D702.7020605@bluedogs-online.com> <20040303014655.69383.qmail@web13121.mail.yahoo.com> <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> Message-ID: <20040303152406.GD16693@socrates> On Tue, Mar 02, 2004 at 10:46:41PM -0500, Jim P wrote: > Writing a tutorial would not have helped the non-technical users. > For a non-technical user, having to consult a tutorial is > a failure of the software. Instead software needs to work so > that a non-technical user does not need a tutorial. > > ESR is trying to change the way all OSS software is done, > not just CUPS. The problem is that the people who have the > skills to make the programs, don't understand that their > software is difficult for non-technical people to use. Hear hear! That's the problem, and that's the diagnosis why it continues to be a problem. Linux is of course nothing *but* a communal technie project, so it's easy to lose sight of people -- the vast majority -- who really really really just want a toaster appliance, and do NOT want to know how or why things work as they do. All computers were like this before the advent of the Apple Macintosh in the early 1980s, and many kept it up until the 1990s. People talk as though Apple's contribution was the GUI, but far more important, it seems to me, was their view that the ordinary user also deserved the full-time efforts of R&D (rather than contempt). "The computer for the rest of us" was a great ad slogan, but also a new (better?) approach to the audience for computers... -- Peter King peter.king@utoronto.ca Department of Philosophy 215 Huron Street The University of Toronto (416)-978-3788 ofc Toronto, ON M5S 1A1 CANADA ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://www.colug.net/pipermail/colug/attachments/20040303/8f060d96/attachment.bin From peter.king at utoronto.ca Wed Mar 3 10:27:48 2004 From: peter.king at utoronto.ca (Peter King) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Segfault suggestions? -- SOLVED In-Reply-To: <20040301155221.GB23021@socrates> References: <20040301155221.GB23021@socrates> Message-ID: <20040303152748.GE16693@socrates> Bad memory, in the end. A bank of sticks 64+32+32, and leaving only the 64MB stick in caused LILO to hang after its BIOS test with a crc error. Removed it entirely, and the computer has survived 11 hours of mprime's torture test with no hitches, and an hour or two of normal usage. Why would the 64MB stick (PC 100) cause LILO to hang by itself but allow the computer to boot up and act flaky in combination with the 32+32 sticks? Any ideas? -- Peter King peter.king@utoronto.ca Department of Philosophy 215 Huron Street The University of Toronto (416)-978-3788 ofc Toronto, ON M5S 1A1 CANADA ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://www.colug.net/pipermail/colug/attachments/20040303/7e266b7f/attachment.bin From jep200305 at columbus.rr.com Wed Mar 3 10:44:57 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Flaky memory In-Reply-To: <20040303152748.GE16693@socrates> References: <20040301155221.GB23021@socrates> <20040303152748.GE16693@socrates> Message-ID: <20040303104457.1bbad885.jep200305@columbus.rr.com> Peter King wrote: > Why would the 64MB stick (PC 100) cause LILO to hang by itself but allow > the computer to boot up and act flaky in combination with the 32+32 > sticks? Any ideas? When you are right on the ragged edge of something working, small changes can tip the behavior one way or the other. In your case, adding memory adds a tiny electrical load to the same wires that the 64MB memory uses. Maybe there was some harsh transient on either the power or a signal to the 64MB memory that was bogged down and mellowed out by adding memory. One doesn't know if you have OK memory in a feeble motherboard or vice versa. Not that you can really do anything about it other than swap stuff in and out. It also wouldn't surprise me if touching the bottom of the motherboard where the memory is with your (static discharged) finger or hand might tip the behavior, not that I'm recommending that you do so. I once had a memory board that worked only when I shined light on it. Shining light on the EPROMs slowed them down. Slowing down the outputs mellowed out transients that the power supply and ground were not good enough for. Figuring out root causes of this kind of stuff can be extremely difficult. Jim From brian at bluedogs-online.com Wed Mar 3 11:29:22 2004 From: brian at bluedogs-online.com (Brian Morgan) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040303152406.GD16693@socrates> References: <4044D702.7020605@bluedogs-online.com> <20040303014655.69383.qmail@web13121.mail.yahoo.com> <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> <20040303152406.GD16693@socrates> Message-ID: <404607E2.6060301@bluedogs-online.com> Peter King wrote: > > > >All computers were like this before the advent of the Apple Macintosh in >the early 1980s, and many kept it up until the 1990s. People talk as >though Apple's contribution was the GUI, but far more important, it >seems to me, was their view that the ordinary user also deserved the >full-time efforts of R&D (rather than contempt). "The computer for the >rest of us" was a great ad slogan, but also a new (better?) approach to >the audience for computers... > > > If any one is interested, here are the sites for usability projects (OSS GUI R&D) for both gnome and kde kde: http://usability.kde.org/ gnome: http://developer.gnome.org/projects/gup/ut1_report/report_main.html Has anyone come across similar projects for other window managers? Brian M From rfunk at funknet.net Wed Mar 3 12:24:19 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Segfault suggestions? -- SOLVED In-Reply-To: <20040303152748.GE16693@socrates> References: <20040301155221.GB23021@socrates> <20040303152748.GE16693@socrates> Message-ID: <200403031224.19210.rfunk@funknet.net> Peter King wrote: > Why would the 64MB stick (PC 100) cause LILO to hang by itself but allow > the computer to boot up and act flaky in combination with the 32+32 > sticks? Any ideas? Besides the current load that Jim mentioned, another possibility is that the bad memory ended up at a different address. LILO would use low-addressed memory, so if the bad memory ended up at some high address, LILO wouldn't notice it. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From tom at functionalmedia.com Wed Mar 3 12:44:13 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> Message-ID: <624C7D49-6D3A-11D8-A6B2-00039317745E@functionalmedia.com> > > Writing a tutorial would not have helped the non-technical users. > For a non-technical user, having to consult a tutorial is > a failure of the software. Instead software needs to work so > that a non-technical user does not need a tutorial. > > ESR is trying to change the way all OSS software is done, > not just CUPS. The problem is that the people who have the > skills to make the programs, don't understand that their > software is difficult for non-technical people to use. > > _______________________________________________ This is a good point, I would like to add the following thoughts to the discussion. I do not know if I agree or not. I guess to a point software should be useable. Do you suppose that a threshold exists that as a product is made more user friendly and focuses on the totally ignorant user that the overall functionality of the system declines? I suppose the theory would be this.. Increase in usability beyond a certain threshold results in a decrease in functionality. A certain level of knowledge is required to do complex things. Printing is a complex thing. Could cups be easier to use yes I suppose. I love CUPS but I agree with ESR's rant, I guess I am calling for a standard knowledge base for usability. I teach an xml class, some of the participants do not understand what a network is. No big deal I can explain what a network is. But it would be easier if everyone new what a network was before they studied XML. It takes 10 minutes to explain what a network is, in basic terms. So we should give our users some knowledge before we throw them to the lions, (the lions here could be printer config, firewall config, connecting to the internet, etc. ) Where MS-windows has failed is that in order to increase usability they start throwing features out the window, or they obfuscate advanced configuration in favor of easy standardized configuration. For example I failed to find the IP's of the dns servers on a win 98 box connected via dhcp. Does an average user need to know the dns servers?? Not usually but sometimes you need to know this information. This is an example of oversimplification. So I think that the tutorial should be mixed into the configuration tool, a help button or a wizard that asks questions while explaining concepts. Or the most basic understanding of computing should be assumed for all users of the system. This "basic knowledge base" should be in every linux-for-newbies distro. Just my 2 cents but I do believe that people are smarter that windows gives them credit for. Give these people a basic grasp of computing and they will be much happier in the long run. Also then when we build user interfaces we can use pieces of this document. When I tell people what happens when they "click on the butterfly" and connect to MSN and what the browser does they are grateful to finally understand what they are doing. Something like the following. -something like this- ++++++++++++++++++++++++++++ Welcome to linux. Linux is an advanced Operating system. It can do many complex things. In order for you to get the most out of this system it is assumed that you have a basic understanding of the following principles. From here the user should be made to understand that his computer has a hard drive, that applications run within the shell. That the window manager runs within the shell. That his/her documents and mp3's are stored on the filesystem. The filesystem is stored on one or more hard drives. The use of the internet involves connecting your machine to a network . Once connected to a network machines can communicate via different protocols. Most importantly computers use numbers to communicate and people tend to use names. So www.google.com is a name. That machine has a number. Think of a phone book. I call Tom Hanlon not by dialing "tom hanlon" but by looking up his phone number. THe phone network and the internet have some similarities, at a basic level. etc etc. Printing: Printing is a complex task. Honestly I do not know how to begin to present an overview of printing. I would make the common techie mistake of going to deep and talking about EPS and spools at the beginning. ++++++++++++++++++ My point is that the complexity should not be hidden too much. The user should be respected as being capable of understanding the basics. The basics are understandable to most folk if the correct language is used. The tools should then be aimed at the user who understands the basics. -the internet is NOT on there computer- -resources documents programs and the like are stored on the hard drive- -etc- I would help prepare such a document. I would need help on some parts. A guide for newbies with the goal of giving them the basics in less than an hour. Before you learn to fly you go to ground school and you learn how flying works. Before you use linux you should at least know what computing is. We could even call the tutorial "ground school for linux pilots" Tom Tom Hanlon Functional Media tom@functionalmedia.com www.functionalmedia.com From JBalint at alldata.net Wed Mar 3 14:10:23 2004 From: JBalint at alldata.net (Balint, Jess) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] To need a tutorial is a failure of the software Message-ID: <5789A42DD684734298BF9EC852D35111033AB26F@scad1exis02.alldata.net> > -the internet is NOT on there computer- Ha ha ha, I love this one.... ~ From McHenryTech at wideopenwest.com Wed Mar 3 20:03:40 2004 From: McHenryTech at wideopenwest.com (Sean McHenry) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Article on Linux as Multimedia PC In-Reply-To: <200403021247.10753.rfunk@funknet.net> References: <200402292351.58732.rfunk@funknet.net> <200403021155.16033.rfunk@funknet.net> <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> <200403021247.10753.rfunk@funknet.net> Message-ID: <1078362220.2373.3.camel@linux.local> You folks might like this one. About fast start up times using Linux as the OS for multimedia PCs. http://www.newscientist.com/news/news.jsp?id=ns99994567 -- Sean McHenry McHenry Technologies Columbus, Ohio Linux Used Here, Since 1998 From roggenkamps at acm.org Wed Mar 3 22:09:49 2004 From: roggenkamps at acm.org (Steve Roggenkamp) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] Article on Linux as Multimedia PC References: <200402292351.58732.rfunk@funknet.net> <200403021155.16033.rfunk@funknet.net> <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> <200403021247.10753.rfunk@funknet.net> <1078362220.2373.3.camel@linux.local> Message-ID: <40469DFD.5090509@acm.org> Check out www.linuxbios.org. They claim a THREE second boot time as their fastest time. Not sure if that's to a console prompt or not though. Here are the first two questions from their FAQ: What is LinuxBIOS? LinuxBIOS aims to replace the normal BIOS found on PCs, Alphas, and other machines with a Linux kernel that can boot Linux from a cold start. LinuxBIOS is primarily Linux - about 10 lines of patches to the current Linux kernel. Additionally, the startup code - about 500 lines of assembly and 5000 lines of C - executes 16 instructions to get into 32-bit mode and then performs DRAM and other hardware initialization required before Linux can take over. Our primary motivation for the project was maintenance of large clusters, but not surprisingly interest and contributions have come from people with varying backgrounds. Why do we need LinuxBIOS? Current PCs used as cluster nodes depend on a vendor-supplied BIOS for booting. The BIOS in turn relies on inherently unreliable devices such as floppy disks and hard drives to boot the operating system. In addition, current BIOS software is unable to accommodate non-standard hardware making it difficult to support experimental work. The BIOS is slow and often erroneous and redundant and, most importantly, maintenance is a nightmare. Imagine walking around with a keyboard and monitor to every one of the 128 nodes in a cluster to change one BIOS setting. The LinuxBIOS gunzip's the Linux kernel straight out of NVRAM and essentially requires no moving parts other than the fan. It does a minimal amount of hardware initialization before jumping to the kernel start and lets Linux do the rest. As a result, it is much faster (current record 3 seconds), which has sparked interest in the consumer electronics community as well. Moreover, updates can be performed over the network. Using a real operating system to boot another operating system provides much greater flexibility than using a simple netboot program or the BIOS. Because Linux is the boot mechanism, it can boot over standard Ethernet or over other interconnects such as Myrinet, Quadrics, or SCI. It can use SSH connections to load the kernel, or it can use the InterMezzo caching file system or traditional NFS. Cluster nodes can be as simple as they need to be - perhaps as simple as a CPU and memory, no disk, no floppy, and no file system. The nodes will be much less autonomous thus making them easier to maintain. Sean McHenry wrote: >You folks might like this one. > >About fast start up times using Linux as the OS for multimedia PCs. > >http://www.newscientist.com/news/news.jsp?id=ns99994567 > > > From jonadab at bright.net Wed Mar 3 23:26:32 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:38 2005 Subject: [COLUG] For all of us who ever used CUPS In-Reply-To: <40455E3C.1050601@bugs.osu.edu> References: <20040303014655.69383.qmail@web13121.mail.yahoo.com> <40455E3C.1050601@bugs.osu.edu> Message-ID: Wade Pinkston writes: > Am I mistaken or is the configuration software not a cups piece of > software. Isn't it a RedHat piece of soft ware? That is possible; I know that Mandrake has its own printerdrake (which on the whole is not bad, though it has its rough edges and problems, some of which are somewhat similar to the ones ESR ranted about), but I don't know what Fedora uses. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Wed Mar 3 23:22:49 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <624C7D49-6D3A-11D8-A6B2-00039317745E@functionalmedia.com> References: <624C7D49-6D3A-11D8-A6B2-00039317745E@functionalmedia.com> Message-ID: tom hanlon writes: > I do not know if I agree or not. I guess to a point software should > be useable. Do you suppose that a threshold exists that as a product > is made more user friendly and focuses on the totally ignorant user > that the overall functionality of the system declines? Only if it's done improperly. Poweruser and even developer functionality *needs* to be present, of course, but it doesn't have to be hanging out in the open by default. Powerusers will explore the menus and find the preferences dialogs. If they see a button labelled "advanced", they'll click it and look at the options in the dialog box that it brings up. End users as a rule will not do these things. So, you cater the out-of-the-box default state as much toward the end user as you can (without seriously compromising security), and then you make it easy to turn on the somewhat more advanced stuff with a few clicks in the prefs dialog. The heavy-duty poweruser stuff you bury in treeview widgets accessed by clicking "Advanced" buttons in the prefs dialog, and then you put links in the prefs and in the help pointing to a page that explains the developer features (such as scriptability, creating your own theme, and such) and how to get the latest bleeding-edge version, get the source, get involved, &c. > I suppose the theory would be this.. Increase in usability beyond a > certain threshold results in a decrease in functionality. It doesn't need to. Unless you count the existence of a "Preferences" option in the menus as a decrease in usability; if the default toolbars are properly designed, most end users will never explore the menubar anyway. One controversial thing that I think should be done is turning off context menus by default and making both mouse buttons do the same thing. It needs to be very easy to change this in the prefs, but the really low-end end users get very confused about which mouse button is which and never need the functionality of most context menus. Ideally, this is the sort of pref that ought to be system-wide and apply to all applications (well, all applications that might conceivably ever be used by end users; a debugger for example could feel free to ignore the global user-experience-level settings and cater itself straight to developers). > A certain level of knowledge is required to do complex > things. Printing is a complex thing. Printing per se is only complex if you're particular about what kind of results you want. Most end users aren't. They don't care how many pages it is or what the margins are or anything; they just want it on paper. (This is a generalization, but it's a _good_ generalization.) _Networked_ printing however is admittedly a little more inherently complex than regular printing, and could require somewhat more knowledge, so setting that up could reasonably ask questions such as, "Is the printer attached to this computer, or another computer?", and in the latter case obviously you've got to find out which printer (though some scanning of the local network for possibilities is warranted). > So we should give our users some knowledge before we throw them to > the lions, (the lions here could be printer config, firewall config, Firewall config should not be done by end users. Poweruser knowledge (at least) is required to do a remotely decent job of it. End users at most should have a switch to flip that turns the stock firewall off or on, and _maybe_ another couple of switches for letting certain kinds of traffic through ("Allow people on the internet to access the web files in your WebFiles folder (Y/N)" -- this could turn Apache off/on, in addition to blocking/unblocking ports 80 and 443) or sharing the internet connection with other local systems (IP Masq or the equivalent). If they want to configure a firewall with any kind of real details, they should be setting their global user experience level up to PowerUser (which incidentally should also have other effects too; individual applications could read this setting to see what menu items to hide or show, and other stuff). > connecting to the internet, etc. ) Where MS-windows has failed is > that in order to increase usability they start throwing features out > the window, Agreed, wrong approach. You don't *remove* features. You hide them unless the user flips a preference that says, "Show me extra options" or whatever. The more advanced the feature is, the bigger the hoop you make the user jump through to turn it on -- slightly advanced features you tuck away in a menu, rather than having a toolbar button. PowerUser features you hide unless a switch is flipped in the prefs dialog, or unless the global user experiece level is turned up to PowerUser. Heavy-duty poweruser features you bury under an Advanced button in the prefs dialog. Only features that really inherently require developer knowledge should require the command line or editing a config file. > or they obfuscate advanced configuration in favor of easy > standardized configuration. An excellent example of this is the MSIE options dialog. If you look at the changes that dialog has gone through over the versions, it has become harder and harder and harder to use in any meaningful way, because none of the options say in plain language what they do; it's all couched in pseudo-easy language that's ambiguous and unclear to powerusers and yet still won't make configuration possible for an end user. You want to enable all cookies? Hahaha. In MSIE4 that was easy. In 5.5 and later you have to know the secret formula. This is the wrong approach. End users shouldn't have to be opening that prefs dialog in the first place, so it should be set up to clearly state what each option does in language that powerusers can understand. > So I think that the tutorial should be mixed into the configuration > tool, a help button or a wizard that asks questions while explaining > concepts. Up to a point, yes. Pointers into the help are good only if the help is helpful and explains things. (Some Windows apps are really good about this; others suck gravel. The ultimate example of useless help is the help for Windows itself, that you find in the Start menu; it's so bad that it should just be removed.) Wizards are an interesting and useful idea, but some things about them as they are currently usually done need to change. Making the user click "Next" six times is needlessly user-hostile, especially if the user is only actually entering three pieces of information. NEVER should there be a dialog where the *only* thing the user needs to do is click on "Next" -- that's useless and annoying. However, the basic idea of Wizards -- asking the user for individual pieces of informatin when they are needed -- is very sound. For example, a mailreader can use this approach to ask the user for their email address and thus avoid the need for the end user to open up the prefs/config dialog at all. (More advanced users will want to set things like signatures, how often to check for new mail, whether to notify with a sound, and so on, so they'll dig into the prefs.) Also, when the app asks the user for the email address, it can then ask something along the lines of, "Do you want me to try to automatically discover your mail server(s) and other details? [Yes, use my email address to figure this stuff out] [No, let me type in the setting my ISP gave me.]". In the former case, it can take the domain from the address, do an MX lookup, and voila, the SMTP server is probably set up correctly (except for auth -- more on that momentarily). Then it can guess that the same computer that's the SMTP server might also be the mailbox server and test whether it's listening for POP3 and/or IMAP connections. So, guess that the username is the same as the username portion of the address, ask the user for a password (this is unavoidable, I'm afraid, and you probably need a "Remember this" checkbox), and try USER/PASS authentication. If USER/PASS works, try to send a message via SMTP to an autoresponder; if you get back the response, mark the account for prior-check authentication and you're done. If you can't send it or don't get the response, try SMTP auth using the POP3 username and password; if _that_ works, mark the account for that and you're done. And so on. There are some snags, things that if they arise you have to ask the user additional questions. For example, if the server supports both POP3 and IMAP, you have to ask the user, "Do you want to keep your mail in folders on the mailserver after you check it, or move it to your own computer?", or something like that. (If the user experience level is set anywhere above the default, you put the initials "IMAP" and "POP3" in parens with each respective option.) Of course, if the user says, "No, let me type in the settings my ISP gave me", then you present them with a fill-in-the-blanks form like most mail clients do in the account settings dialog. Mail client configuration is just an example. The key point is that if the user hasn't set a preference to indicate that they're at least somewhat knowledgeable, you only ask them what you absolutely can't avoid asking, and you never make them dig into prefs dialogs. > Just my 2 cents but I do believe that people are smarter that > windows gives them credit for. Many of them can be, if properly coached, but you want to break them in somewhat gradually. > Give these people a basic grasp of computing and they will be much > happier in the long run. Yes, but you don't want them to have to get their basic grasp the first day before they can do anything. You want to teach it to them gradually. One think I like is a tips dialog that cycles through a tips file, getting a new tip each time the app starts (and starting over after the end), with a checked-by-default "Continue showing a tip each time" checkbox. One new tip each day is just about what a lot of people can hope to absorb, especially if they're properly ordered. Sure, some of the tips they won't care about, but others will help, and they're easy enough to turn off. Ideally, this sort of thing should be sensitive to the global user experience level setting and not show tips that are significantly below or above the user's level. > My point is that the complexity should not be hidden too much. Not too much, but not too little either. You want the complexity just beneath the surface, where it's easy to get to but doesn't hit end users in the face all the time. Also, you want to gradate it and put the more serious heavy-duty complexity further under the surface. > The user should be respected as being capable of understanding the > basics. Capable of understanding, yes -- already knowing it all, no. This is the "discoverability" that ESR was talking about. > The basics are understandable to most folk if the correct language > is used. The tools should then be aimed at the user who understands > the basics. Many tools can be aimed at both levels. > -the internet is NOT on there computer- Heh. People ask me if I have Yahoo on my computer. I make a point of always telling them that Yahoo is in Southern California. (Then after that sinks in for a couple of seconds I answer the question I think they meant to ask -- e.g., do I access Yahoo from my computer, or am I familiar enough with Yahoo to answer a question about it.) So yes, I agree with you that an attempt should be made to (gently) educate users, but I think it's also important to do that a little at a time, at a level where they can understand it given what they already know. It's not that users are stupid; it's just that they've spent their lives doing other stuff and so don't know a lot about computers. You can teach them bits and pieces, but you can't assume that they already know stuff. Also, you usually can't assume that they're going to sit down and read a couple of pages at one sitting and learn all of it. People think they're busy (even if they spend six hours a day watching movies and television, they think they're "busy"), so there's a limit to how much they usually want to absorb at one time. There are exceptions; sometimes you can get people to sign up for a two-hour course in internet, word processing, or whatever. (Part of my job at the GPL is to teach these.) But at other times, when they're in a hurry, you have to give them a smaller amount that they're willing to ingest and let the rest go until later. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Wed Mar 3 23:41:38 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB26F@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB26F@scad1exis02.alldata.net> Message-ID: "Balint, Jess" writes: > > -the internet is NOT on there computer- > > Ha ha ha, I love this one.... You laugh; this is a useful point to make with end users. They theoretically know better, but they tend to think of the internet as being on their computer, because that's where they experience it. It is also notable that to many end users, the following terms are all mostly interchangeable: "CPU", "modem", "hard drive". They all refer to the case. (Vendors do NOT help with this when they sell the whole computer as a "CPU", which is annoyingly common.) This one I don't usually bother to dispel, other than to consistently refer to it myself as "the computer" (always placing my hand on it when doing so, because otherwise there is a danger that some of them might think I mean the monitor; I always call the monitor the "screen" or "display"). If it's a tower case, you can call it "the tower" and the end user will know what you mean. The term "case" is likely not to be understood, however. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From chris at chrisbaker.net Thu Mar 4 00:10:42 2004 From: chris at chrisbaker.net (Christopher Baker) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] Fw: [FreeOhio] Cincinnati power company begins high-speed ISP service Message-ID: <20040304001035.596F.CHRIS@chrisbaker.net> This isn't a Linux issue per se, but I think people here will find it intriguing. Tuesday, March 2, 2004 Cinergy plugs in as Web provider Power lines set to carry Internet to outlet near you http://www.enquirer.com/editions/2004/03/02/biz_biz1acin.html By Mike Boyer The Cincinnati Enquirer A unit of Cinergy Corp. today will become the nation's first electric utility to offer high-speed Internet service to customers via its power lines, turning every electrical outlet in homes or offices into a Web connection. The technology, which will be offered first in Hyde Park and Mount Lookout, holds the promise of adding competition and cutting prices for broadband services while making such service available (particularly in remote and rural areas) without costly investments in cables...... E-mail mboyer@enquirer.com Distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. ============================================================ FRANK BURNS: Unless we each conform, unless we obey orders, unless we follow our leaders blindly, there is no possible way we can remain free. _M*A*S*H_, "The Novocaine Mutiny" ============================================================ Chris Baker -- www.chrisbaker.net chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com "When you stop growing, you start dying." From anderson at wks.uts.ohio-state.edu Thu Mar 4 09:39:20 2004 From: anderson at wks.uts.ohio-state.edu (Bill Anderson) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] 3490E tape drive Message-ID: If anyone happens to have access to a tape drive which can read 3490E 36-track tapes, or knows where I could find one cheap, please mail me off list. Thanks, Bill From kb8rln at penguinmaster.com Thu Mar 4 10:18:33 2004 From: kb8rln at penguinmaster.com (Richard Rager) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] Fw: [FreeOhio] Cincinnati power company begins high-speed ISP service In-Reply-To: <20040304001035.596F.CHRIS@chrisbaker.net> References: <20040304001035.596F.CHRIS@chrisbaker.net> Message-ID: On Thu, 4 Mar 2004, Christopher Baker wrote: > Tuesday, March 2, 2004 > > Cinergy plugs in as Web provider Power lines set to carry Internet to outlet > near you > > The technology, which will be offered first in Hyde Park and Mount Lookout, > holds the promise of adding competition and cutting prices for broadband > services while making such service available (particularly in remote and rural > areas) without costly investments in cables...... > So much for HF radio. :-( de kb8rln Richard Rager From vrman at bourbaki.us Thu Mar 4 10:45:55 2004 From: vrman at bourbaki.us (Richard J. Mancusi) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] Fw: [FreeOhio] Cincinnati power company begins high-speed ISP service In-Reply-To: <20040304001035.596F.CHRIS@chrisbaker.net> References: <20040304001035.596F.CHRIS@chrisbaker.net> Message-ID: <40474F33.5070804@bourbaki.us> You may find this link interesting: http://www.arrl.org/tis/info/HTML/plc/ Rich Christopher Baker wrote: > This isn't a Linux issue per se, but I think people here will find it > intriguing. > > Tuesday, March 2, 2004 > > Cinergy plugs in as Web provider Power lines set to carry Internet to outlet > near you > From nliu99 at yahoo.com Thu Mar 4 13:41:57 2004 From: nliu99 at yahoo.com (Nan Liu) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <200403041700.i24H0Qs3028549@stones.wcbe.org> Message-ID: <20040304184157.6744.qmail@web80601.mail.yahoo.com> > "Balint, Jess" writes: > > > > -the internet is NOT on there computer- > > > > Ha ha ha, I love this one.... Not nearly as bad as someone who asked me:"I have phone service, so why can't I get on the internet?" It was a nightmare trying to explain to her what an ISP was, and why there was additional charges. > You laugh; this is a useful point to make with end users. They > theoretically know better, but they tend to think of the internet as > being on their computer, because that's where they experience it. > I'm OK with the users being ignorant. I don't expect my parents to become a sysadmin or even a guru before they understand how to use a computer. But I'm sick of bestbuy/compusa/fill-in-the-blank wannabe "techies" telling customers flat out lies such as "upgrade your memory for faster internet experience", "2.6 GHZ Pentium (umm celerons) are faster than AMD 2.5". They charge an extortion and they don't fix the problems. Case and point, BB ran diagnostic and sold my parents more RAM, when the computer slowdown was caused by a virus. The sales people are twice as bad, they just throw out random technical terms: hyperthreading ("you can do twice as much!"), ghz("faster is better!"), wireless to impress the customers. I was about to laugh my head off when one of the sales people tried to tell a customer:"Sorry we only have minus CDR (label says CD-R), but we have plus DVD+R". When I told him CDR are universal, only DVDs have +/- standard, he started rambling:"The information is in our training guide! See for yourself!" Umm, no thanks. Another example, customer asking what's a VOB file on the Sony DVD-R camcorders, and how can he edit the video, sales person just rambled:"I think that's a quicktime or windows media format, if you have XP media center edition you should be able to edit it." With all these idiots feeding mis-information, I'd say most users don't know better, otherwise these places would be out of business years ago. Before we start educating the users, we seriously need to replace these idiots giving them bad information. From jsinglet at insight.rr.com Thu Mar 4 16:57:19 2004 From: jsinglet at insight.rr.com (Jim Singleton) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] Fw: [FreeOhio] Cincinnati power company begins high-speed ISP service In-Reply-To: References: <20040304001035.596F.CHRIS@chrisbaker.net> Message-ID: <1078437439.21896.4.camel@linux.local> On Thu, 2004-03-04 at 10:18, Richard Rager wrote: > On Thu, 4 Mar 2004, Christopher Baker wrote: > > > Tuesday, March 2, 2004 > > > > Cinergy plugs in as Web provider Power lines set to carry Internet to outlet > > near you > > > > The technology, which will be offered first in Hyde Park and Mount Lookout, > > holds the promise of adding competition and cutting prices for broadband > > services while making such service available (particularly in remote and rural > > areas) without costly investments in cables...... > > > > So much for HF radio. :-( > > > de kb8rln > > Richard Rager Whats going to be more interesting is whats going to happen to all those users when some one keys up on 10/20m (and up) w/1000 watts, more than a few users are going to loose their connection. Which should make an interesting complaint to the FCC. Considering were the primary users on those bands.Not all is lost but it's going to get interesting. de kc8wif Jim From vherried at insight.rr.com Thu Mar 4 18:06:02 2004 From: vherried at insight.rr.com (Vince Herried) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] yum users beware. Message-ID: <200403041806.09693.vherried@insight.rr.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 It appears that one of these updates broke yum. libxml2-0-2.5.4-3.rh9.i388.rpm libxml2-python-0-2.5.4-3.rh9.i386.rpm libxml2-devel-0-2.5.4-3.rh9.i386.rpm since these updates, yum generates a bunch of diagnostic messages no matter what parms are given. - -- Vince Herried GPG keyID 0xD004D08366 (8366 F97C AE0A 59AA CB14 952D 85C6 304A D004 D001) gpg --keyserver pgp.mit.edu --recv-keys D004D001 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFAR7ZhhcYwStAE0AERAr7tAKCwwShhMTVEFLSftyhI8CnXChTdOQCgqJEH GKi76vjVCFGkyYDQ8sOnoSI= =ACsa -----END PGP SIGNATURE----- From meffiem at neo.rr.com Thu Mar 4 19:10:54 2004 From: meffiem at neo.rr.com (Michael Meffie) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: References: <5789A42DD684734298BF9EC852D35111033AB26F@scad1exis02.alldata.net> Message-ID: <20040304191054.6fe6fcf0.meffiem@neo.rr.com> > myself as "the computer" (always placing my hand on it when doing so, > because otherwise there is a danger that some of them might think I > mean the monitor; I always call the monitor the "screen" or > "display"). I've see some people refer to a monitor as "the computer" (esp kids). That really shocked me :) From meffiem at neo.rr.com Thu Mar 4 19:39:00 2004 From: meffiem at neo.rr.com (Michael Meffie) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040303152406.GD16693@socrates> References: <4044D702.7020605@bluedogs-online.com> <20040303014655.69383.qmail@web13121.mail.yahoo.com> <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> <20040303152406.GD16693@socrates> Message-ID: <20040304193900.4498e444.meffiem@neo.rr.com> > > ESR is trying to change the way all OSS software is done, > > not just CUPS. The problem is that the people who have the > > skills to make the programs, don't understand that their > > software is difficult for non-technical people to use. Actually, ESR is a technical user. His point was that common tasks should be simple and quick, even for a very busy technical user. His points are very good, especially on the aspect of discoveriability in the configuration process. Discoverability is the configuration process has to be fundemental, and is the main reason people like knoppix so much. > That's the problem, and that's the diagnosis why it continues to be a > problem. Linux is of course nothing *but* a communal technie project, > so it's easy to lose sight of people -- the vast majority -- who really > really really just want a toaster appliance, and do NOT want to know how > or why things work as they do. Linux the kernel, Linux the free/open-source projects on source forge, or Linux the LindowOS distribution? Obviously, those with commercial interests may have the most to gain for this work. It seems to me, to have usable software, for any kind of task, requires the insight on user expectations. Now, if technical people are writing the code for themselves, you'll end up with emacs and vi. But to make software such as a simple paint brush program work well, you have to watch people using it for the first time, and see where they are getting frustrated, and what they do like. Basic software development. I'm not convinced open-source/free software can achieve this. Perhaps the vendors will be able to make that happen, if they can stay afloat. From jep200305 at columbus.rr.com Thu Mar 4 19:54:02 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] yum users beware. In-Reply-To: <200403041806.09693.vherried@insight.rr.com> References: <200403041806.09693.vherried@insight.rr.com> Message-ID: <20040304195402.3c984927.jep200305@columbus.rr.com> Vince wrote: > It appears that one of these updates broke yum. .... > since these updates, yum generates a bunch of diagnostic messages > no matter what parms are given. I've heard comments that Red Hat's yum is not advanced as other's yum. From dshermin at ameritech.net Thu Mar 4 20:32:31 2004 From: dshermin at ameritech.net (David Sherman) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] SCO vs Bank of America Message-ID: http://news.com.com/2100-7344-5170073.html From whooper at freeshell.org Thu Mar 4 20:49:35 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] yum users beware. In-Reply-To: <20040304195402.3c984927.jep200305@columbus.rr.com> References: <200403041806.09693.vherried@insight.rr.com> <20040304195402.3c984927.jep200305@columbus.rr.com> Message-ID: <64454.69.68.50.13.1078451375.squirrel@69.68.50.13> Jim P said: > Vince wrote: > >> It appears that one of these updates broke yum. > .... >> since these updates, yum generates a bunch of diagnostic messages >> no matter what parms are given. > > I've heard comments that Red Hat's yum is not advanced > as other's yum. That wouldn't be hard, Red Hat doesn't ship yum for RHL9. -- William Hooper From archanoid at columbus.rr.com Thu Mar 4 20:55:53 2004 From: archanoid at columbus.rr.com (Aaron Howard) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: References: Message-ID: On Thu, 04 Mar 2004 20:32:31 -0500, David Sherman wrote: > http://news.com.com/2100-7344-5170073.html > Just out of curiosity, how many people on this list are *not* following www.groklaw.net? -Aaron From skippy at skippy.net Thu Mar 4 21:08:17 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: References: Message-ID: <4047E111.3010709@skippy.net> Aaron Howard wrote: > On Thu, 04 Mar 2004 20:32:31 -0500, David Sherman > wrote: > >> http://news.com.com/2100-7344-5170073.html >> > > Just out of curiosity, how many people on this list are *not* following > www.groklaw.net? Not I. I'm not interested enough in the details to pursue anything other than the headlines as they bounce around the sites I frequent. From whooper at freeshell.org Thu Mar 4 21:04:16 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] yum users beware. In-Reply-To: <200403041806.09693.vherried@insight.rr.com> References: <200403041806.09693.vherried@insight.rr.com> Message-ID: <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> Vince Herried said: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > It appears that one of these updates broke yum. > > > libxml2-0-2.5.4-3.rh9.i388.rpm > libxml2-python-0-2.5.4-3.rh9.i386.rpm > libxml2-devel-0-2.5.4-3.rh9.i386.rpm > > since these updates, yum generates a bunch of diagnostic messages > no matter what parms are given. I'm 99% sure I've done these updates at work and yum hasn't had a problem. What version of yum? Maybe one of your repos is overly busy? -- William Hooper From blata at extent0006.entomology.ohio-state.edu Thu Mar 4 21:10:20 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: <4047E111.3010709@skippy.net> References: <4047E111.3010709@skippy.net> Message-ID: <4047E18C.4040304@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 im not despite not knowing about it. You wrote this to me on 03/04/2004 09:08 PM: | Aaron Howard wrote: | |> On Thu, 04 Mar 2004 20:32:31 -0500, David Sherman |> wrote: |> |>> http://news.com.com/2100-7344-5170073.html |>> |> |> Just out of curiosity, how many people on this list are *not* |> following www.groklaw.net? | | | Not I. I'm not interested enough in the details to pursue anything | other than the headlines as they bounce around the sites I frequent. | _______________________________________________ | colug mailing list | colug@colug.net | http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAR+GMv+6+qSFv3TURAnhPAJ9CDCcJwRMqXkvhO4f93ypx3GleYQCfbaB5 v6fP6W/gAyOCET9J+LDacMY= =D+Yb -----END PGP SIGNATURE----- From whooper at freeshell.org Thu Mar 4 21:15:29 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:39 2005 Subject: [COLUG] yum users beware. In-Reply-To: <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> References: <200403041806.09693.vherried@insight.rr.com> <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> Message-ID: <64729.69.68.50.13.1078452929.squirrel@69.68.50.13> William Hooper said: > > Vince Herried said: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> It appears that one of these updates broke yum. >> >> >> libxml2-0-2.5.4-3.rh9.i388.rpm >> libxml2-python-0-2.5.4-3.rh9.i386.rpm >> libxml2-devel-0-2.5.4-3.rh9.i386.rpm >> >> since these updates, yum generates a bunch of diagnostic messages >> no matter what parms are given. > > I'm 99% sure I've done these updates at work and yum hasn't had a problem. > What version of yum? Maybe one of your repos is overly busy? (Now that my home RHL9 machine has caught up after being off for a while...) I'm 100% that it works: [whooper@poit whooper]$ rpm -q libxml2 libxml2-python libxml2-devel yum libxml2-2.5.4-3.rh9 libxml2-python-2.5.4-3.rh9 package libxml2-devel is not installed yum-2.0.4-1.rh.fr [whooper@poit whooper]$ -- William Hooper From brucehohl at yahoo.com Thu Mar 4 21:19:05 2004 From: brucehohl at yahoo.com (Bruce Hohl) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040304193900.4498e444.meffiem@neo.rr.com> Message-ID: <20040305021905.2928.qmail@web13126.mail.yahoo.com> --- Michael Meffie wrote: > > But to make software such as a simple paint brush > program work well, you have to watch people using it > for the first time, and see where they are > getting frustrated, and what they do like. Basic > software development. I'm not convinced > open-source/free software can achieve this. > KDE, OpenOffice.org, Kb3, KOffice, XMMS, etc, etc. I think Open Source can and has delivered much easy to use software. ESR should not generalize. There is plenty of difficult to use commercial software out there. Perhaps the only thing CUPS lacks is a quick set up guide for local printer, SMB printer, etc. __________________________________ Do you Yahoo!? Yahoo! Search - Find what you’re looking for faster http://search.yahoo.com From dmaxwell at columbus.rr.com Thu Mar 4 21:47:32 2004 From: dmaxwell at columbus.rr.com (Dave Maxwell) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: References: Message-ID: <200403042147.32736.dmaxwell@columbus.rr.com> On Thursday 04 March 2004 08:55 pm, Aaron Howard wrote: > On Thu, 04 Mar 2004 20:32:31 -0500, David Sherman > > wrote: > > http://news.com.com/2100-7344-5170073.html > > Just out of curiosity, how many people on this list are *not* following > www.groklaw.net? > I've been paying entirely too much attention. I can boil all of the hullabaloo down to this: When do we get some licks in? IBM is pursuing their usual strategy of defeat-in-detail. In the long term that is for the best but they get to do quite a bit of damage in the meantime. The RedHat suit is exactly the sort of thing I would like to see more of. RedHat is at least trying to take the fight directly to SCO as are others in Germany and Austrailia. It looks for all the world like the Delaware courts are sitting on their hands so the Utah courts can do their job for them; justice delayed can be justice denied. What I'd really like to see is some kernel hackers sticking it right to them. Since SCO loves the DMCA so much, SCO getting their web hosting shut down for a week or two would be a good start on serving them right...and I'm not thinking DDOS. SCO is in outrageous violation of the kernel hackers' copyright. I see nothing wrong with using the enemy's own DMCA against them. I know, I know, the moral high ground but then it was Dark Helmet who reminded us that "Evil will always triumph because Good is dumb." I figure a taste of their own medicine is thoroughly in order. Then the hackers can really get nasty considering the punitive damages for copyright violation alone. If I'm not mistaken, the kernel has patented work in it as well...only if the GPL is honored. There's a legal fund I'd be ecstatic to chip into. Dave -- 1st graffitiist: QUESTION AUTHORITY! 2nd graffitiist: Why? From vherried at insight.rr.com Thu Mar 4 22:09:18 2004 From: vherried at insight.rr.com (Vince Herried) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] yum users beware. In-Reply-To: <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> References: <200403041806.09693.vherried@insight.rr.com> <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> Message-ID: <200403042209.21959.vherried@insight.rr.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday 04 March 2004 09:04 pm, William Hooper wrote: > Vince Herried said: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > It appears that one of these updates broke yum. > > > > > > libxml2-0-2.5.4-3.rh9.i388.rpm > > libxml2-python-0-2.5.4-3.rh9.i386.rpm > > libxml2-devel-0-2.5.4-3.rh9.i386.rpm > > > > since these updates, yum generates a bunch of diagnostic messages > > no matter what parms are given. > > I'm 99% sure I've done these updates at work and yum hasn't had a problem. > What version of yum? Maybe one of your repos is overly busy? yum-2.0.4-1.rh.fr I get [root@lap0 root]# yum -help Traceback (most recent call last): File "/usr/bin/yum", line 22, in ? import yummain File "/usr/share/yum/yummain.py", line 21, in ? import clientStuff File "/usr/share/yum/clientStuff.py", line 25, in ? import pkgaction File "/usr/share/yum/pkgaction.py", line 25, in ? import rpmUtils File "/usr/share/yum/rpmUtils.py", line 9, in ? from urlgrabber import URLGrabError File "/usr/share/yum/urlgrabber.py", line 21, in ? import urllib2 File "/usr/lib/python2.2/urllib2.py", line 101, in ? import ftplib File "/usr/lib/python2.2/ftplib.py", line 68, in ? all_errors = (Error, socket.error, IOError, EOFError) AttributeError: 'module' object has no attribute 'error' [root@lap0 root]# on another machine w/o the updates it works ok. Vince - -- GPG keyID 0xD004D08366 (8366 F97C AE0A 59AA CB14 952D 85C6 304A D004 D001) gpg --keyserver pgp.mit.edu --recv-keys D004D001 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFAR+9hhcYwStAE0AERAgtXAJ90cMsJs5VLLDmFePTpw6hNtPD/eQCfUUqY yoMzdNunJbUT4W8N1gWO3Qw= =g+kK -----END PGP SIGNATURE----- From mwmiller at columbus.rr.com Fri Mar 5 04:23:52 2004 From: mwmiller at columbus.rr.com (Matthew W. Miller) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: <200403042147.32736.dmaxwell@columbus.rr.com> References: <200403042147.32736.dmaxwell@columbus.rr.com> Message-ID: <20040305092352.GA23257@columbus.rr.com> On Thu, Mar 04, 2004 at 09:47:32PM -0500, Dave Maxwell wrote: >SCO is in outrageous violation of the kernel hackers' copyright. I see >nothing wrong with using the enemy's own DMCA against them. In my humble opinion, this is not a good idea. Aside from taking "the moral high ground", I do not want to lend any credence to the DMCA being a constitutional and enforceable law, period. A law that's a pain in the ass for the "bad guys" can (and, in my cynical view, will) be a pain in the ass for the "good guys" as well. -- Matthew W. Miller MPlayer does not suck. From beano21 at direcway.com Fri Mar 5 07:11:08 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: Message-ID: I would like to provide my 2cents.(Yes I go to Groklaw) I USE Linux. Not for Business, nor to make money. A big Company throwing its perverbial wieght around just makes me wonder why all of a sudden is SCO wants to get to USERS of Linux. We didn't violate copyright by coding something. That is kinda like giving the passenger in a car a ticket because the driver was speeding. I did however take a step back and look at somethings. Microsoft is not in this fight on the frontline. Why is that? Who ever sends a Word Document? I work at a bank and anything external is supposed to be PDF'd and Encrypted. I would dare to say security for this firm is as bad as their clints. I also see a few things happening. SCO has made Linux a houshold name. I got a call from my Aunt in New York. She has never been and still is not good with computers. Because she remembered I use linux she gave me a call to see if I was being sued and what I though about SCO. I am blind to anything but Linux coming out of this lawsuit without a scratch. Overall this is good for Linux. Once this is over Linux can only get bigger. I also think any company that paid the fee will sue SCO. This will end with either a hit to linux or SCO run out of business. Thank You. -----Original Message----- From: colug-bounces@colug.net [mailto:colug-bounces@colug.net]On Behalf Of David Sherman Sent: Thursday, March 04, 2004 8:33 PM To: colug@colug.net Subject: [COLUG] SCO vs Bank of America http://news.com.com/2100-7344-5170073.html _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From whooper at freeshell.org Fri Mar 5 08:25:26 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] yum users beware. In-Reply-To: <200403042209.21959.vherried@insight.rr.com> References: <200403041806.09693.vherried@insight.rr.com> <64700.69.68.50.13.1078452256.squirrel@69.68.50.13> <200403042209.21959.vherried@insight.rr.com> Message-ID: <2616.12.29.16.103.1078493126.squirrel@12.29.16.103> Vince Herried said: > > yum-2.0.4-1.rh.fr Now that I'm at work, I can confirm that one works also: [whooper@testweb whooper]$ rpm -q libxml2 libxml2-python libxml2-devel yum libxml2-2.5.4-3.rh9 libxml2-python-2.5.4-3.rh9 libxml2-devel-2.5.4-3.rh9 yum-2.0.4-1.rh.fr [whooper@testweb whooper]$ > I get > > [root@lap0 root]# yum -help > Traceback (most recent call last): [snip] > all_errors = (Error, socket.error, IOError, EOFError) > AttributeError: 'module' object has no attribute 'error' > [root@lap0 root]# I'm afraid I haven't seen those errors before. It would probably be best to post to the FreshRPMS list (http://freshrpms.net/lists/) and see if anyone can help. -- William Hooper From colug at jmglov.net Fri Mar 5 09:59:12 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040305021905.2928.qmail@web13126.mail.yahoo.com> References: <20040304193900.4498e444.meffiem@neo.rr.com> <20040305021905.2928.qmail@web13126.mail.yahoo.com> Message-ID: <1158.24.123.50.150.1078498752.squirrel@mail.jmglov.net> Quoth Bruce Hohl: > I think Open Source can and has delivered much easy to use software. ^^^^ some The problem is that you are a technical user. You go to man pages, search Google, etc, to get something to work. The average non-technical user is unlikely to use any online help. That is the difference. My wife is a non-technical user. I watch how she uses her GNOME desktop, and am often surprised at the things that frustrate her. This is the kind of research that Apple does. This is the kind of research that others do not. Were I interested in building software that is discoverable to non-technical users, I would watch how my wife, my parents, my grandparents respond to unfamiliar software. Luckily for me, the software that I write is aimed at more technical users. > ESR should not generalize. Why not? Generalisations are a useful tool as long as they are not abused. The fact that the CUPS developers responded positively, as did most Open Source developers that commented on his rant on /. et al, seems to indicate to me that the shoe does fit, largely. > There is plenty of difficult to use commercial software out there. Yes, there is. That was not ESR's point, however. If you are talking about building a car, you should not make the argument that there are plenty of dangerous cars out there. You should make your car safer, so more people will buy it. > Perhaps the only thing CUPS lacks is a quick set up guide for local printer, > SMB printer, etc. Which no non-technical reader will read. No, as ESR says, the wizard must be self-documenting. And as he pointed out, in the case of the Red Hat / CUPS add printer wizard, they almost got it right. When your target audience is the non-technical user (or the technical user who cannot or will not research your program), as you seem to indicate by choosing a wizard interface, you need to provide sensible defaults and brief verbiage that directs the user through the common cases that will cover 90% of users' requirements. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 From lefevre.10 at osu.edu Fri Mar 5 15:25:59 2004 From: lefevre.10 at osu.edu (Steve Lefevre) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <1158.24.123.50.150.1078498752.squirrel@mail.jmglov.net> References: <20040304193900.4498e444.meffiem@neo.rr.com> <20040305021905.2928.qmail@web13126.mail.yahoo.com> <1158.24.123.50.150.1078498752.squirrel@mail.jmglov.net> Message-ID: <4048E257.5020304@osu.edu> Josh Glover wrote: >Which no non-technical reader will read. No, as ESR says, the wizard must be >self-documenting. And as he pointed out, in the case of the Red Hat / CUPS add >printer wizard, they almost got it right. > > To me, that's the definition of a wizard. It's a documented walk-through of the settings, forcing you to make decisions, and skipping irrelevant settings based on previous decisions. Steve Lefevre From jonadab at bright.net Fri Mar 5 23:45:05 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040304193900.4498e444.meffiem@neo.rr.com> References: <4044D702.7020605@bluedogs-online.com> <20040303014655.69383.qmail@web13121.mail.yahoo.com> <20040302224641.5bd7e1fb.jep200305@columbus.rr.com> <20040303152406.GD16693@socrates> <20040304193900.4498e444.meffiem@neo.rr.com> Message-ID: Michael Meffie writes: > It seems to me, to have usable software, for any kind of task, > requires the insight on user expectations. Now, if technical people > are writing the code for themselves, you'll end up with emacs and > vi. Actually, Emacs has most of the makings of a great text editor for lower-end powerusers. (Real end users, of course, don't need a text editor except the one built into their mail client and the textarea widget of their browser. But the Emacs learning curve scares off some people who are sufficiently advanced to regularly use e.g. Notepad, and that is a shame.) It just (though this is admittedly a large "just") needs to have a bunch of defaults changed (mostly keybindings, menus, and stuff). This could be done in a big fat elisp module. (It would be more than just a bunch of calls to setq and global-set-key, though, because it would have to set hooks for a lot of modes, among other things.) I've seriously thought about doing this, but I concluded that because of the large number of things that would need to be rebound I would not be able to keep myself motivated to finish it after the functions I use regularly had been rebound. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Fri Mar 5 23:52:24 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040304191054.6fe6fcf0.meffiem@neo.rr.com> References: <5789A42DD684734298BF9EC852D35111033AB26F@scad1exis02.alldata.net> <20040304191054.6fe6fcf0.meffiem@neo.rr.com> Message-ID: Michael Meffie writes: > > myself as "the computer" (always placing my hand on it when doing so, > > because otherwise there is a danger that some of them might think I > > mean the monitor; I always call the monitor the "screen" or > > "display"). > > I've see some people refer to a monitor as "the computer" (esp > kids). That really shocked me :) Yep. That's why when I say "the computer" to an end user I rest my hand on the case to clarify that this is the object I mean. OTOH, when I say "the screen", everybody understands which part I mean. Also, "the keyboard" and "the mouse" communicate precisely. Sometimes there's a little trouble with mouse-related terms, though. An especially common misconception is that "click" means to press the button down, which leads confusion over the difference between clicking versus dragging. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From 6f at earthlink.net Fri Mar 5 23:57:23 2004 From: 6f at earthlink.net (Steve Ferguson) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] Sound Message-ID: <20040306045723.GB5485@earthlink.net> Howdy- I'm having some sound issues. Here is the text on the audio from dmesg: via 686a/8233/8235 audio driver 1.9.1-ac2 via82cxxx: six channel audio driver I can't get any audio to work with XMMS (or anything else for that matter). I've tried all combinations of plugins, etc. Any ideas? Anyone work with this driver before? Thanks, Steve From jonadab at bright.net Sat Mar 6 00:12:12 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040304184157.6744.qmail@web80601.mail.yahoo.com> References: <20040304184157.6744.qmail@web80601.mail.yahoo.com> Message-ID: Nan Liu writes: > > You laugh; this is a useful point to make with end users. They > > theoretically know better, but they tend to think of the internet as > > being on their computer, because that's where they experience it. > > I'm OK with the users being ignorant. Up to a point, but serious misconceptions like this one have a tendency to have a negative impact on their ability to use the computer (or, in this case, the internet). > I don't expect my parents to become a sysadmin or even a guru before > they understand how to use a computer. No, no, of course not. I did teach my parents what a web browser is, though. > But I'm sick of bestbuy/compusa/fill-in-the-blank wannabe "techies" > telling customers flat out lies such as "upgrade your memory for > faster internet experience", "2.6 GHZ Pentium (umm celerons) are > faster than AMD 2.5". They charge an extortion and they don't fix > the problems. Case and point, BB ran diagnostic and sold my parents > more RAM, when the computer slowdown was caused by a virus. Ugh. I don't think end users should try to make computer purchase decisions if they can avoid it. If there's nobody in the family who knows about computers, then they're probably stuck, but I sure don't make my parents figure out what computer to buy themselves -- and every time I've slipped and let them make a computer-related purchase decision, I've regretted it. For example, there's my mom's scanner, which she picked out from an advertisement and bought, an HP ScanJet 5470c. That thing's horrible. Well, the quality of the scans is good enough, but the driver/software is so aweful that there's absolutely no prayer anybody in the family but me can scan anything, and it took me the better part of an hour to figure it out the first time, and I curse the interface every time I use it. I wanted to take it back, but my mom saw the result of the scan on the screen after an hour, squealed with delight, clapped her hands, ... (okay, I'm exaggerating, but she was just happy that it "worked" and didn't think a painfully bad interface was a good reason to return it; it's probably my fault for figuring out how to make it work; if I'd given up in disgust after fifteen minutes, she'd probably have agreed to return the beast). > The sales people are twice as bad Most of the people who sell computers are _salesmen_, not computer people. They don't know significantly more about the computers than the people they're selling them to. (Sometimes I think the ones at Radio Shack know _less_ about computers than the people they're selling them to.) > With all these idiots feeding mis-information, I'd say most users > don't know better, otherwise these places would be out of business > years ago. Before we start educating the users, we seriously need > to replace these idiots giving them bad information. Not likely. They're _salesmen_. If they make a sale, they're not likely going to lose their jobs for lack of also handing out correct technical information. We *might* have some chance of educating the users to ask tech people instead of salesmen when they want accurate information. Well, some users, anyway. I've got my parents trained to ask me first if they have any questions related to higher math or computers. My boss is starting too, to get the idea that I know more about computers than the salesmen at our vendor. (The vendor also has tech people, and they know more than I do, especially about VMS, but the sales people don't (well, not about computers).) -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Sat Mar 6 00:19:42 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] SCO vs Bank of America In-Reply-To: References: Message-ID: Aaron Howard writes: > Just out of curiosity, how many people on this list are *not* > following www.groklaw.net? Not in detail. I have a passing knowledge of the basic jist of most of what they cover, but I don't read most of their articles en toto. The details aren't important to me; when the whole thing first came out I read up enough to determine that SCO's claims are fundamentally bogus, and I read enough slashdot to get a general feel for how long the whole fiasco is taking to get itself ironed out, but the specifics change so much from day to day that I don't guess they matter. The end result is pretty predicatable: nothing significant. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From rfunk at funknet.net Sat Mar 6 00:54:43 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: References: <4044D702.7020605@bluedogs-online.com> <20040304193900.4498e444.meffiem@neo.rr.com> Message-ID: <200403060054.43518.rfunk@funknet.net> Jonadab the Unsightly One wrote: > Michael Meffie writes: > > Now, if technical people > > are writing the code for themselves, you'll end up with emacs and > > vi. > > Actually, Emacs has most of the makings of a great text editor for > lower-end powerusers. (Real end users, of course, don't need a text > editor except the one built into their mail client and the textarea > widget of their browser. But the Emacs learning curve scares off some > people who are sufficiently advanced to regularly use e.g. Notepad, > and that is a shame.) Back when I started using email at OSU in 1992, I had a choice between three different editors: emacs, vi, and one I'm pretty sure was a home-brewed (at OSU) hack called fse. With no Unix experience at the time, I chose emacs because it was the only one that seemed to handle my arrow keys properly. (fse was supposed to be simpler, but apparently was hard-coded for vt100 terminals and I was using a vt52 terminal emulator on my Atari.) Learning the right keys to do the basics in emacs was pretty easy, between the on-screen help and the cheat-sheet I got from OSU. The only thing I ever found hard about emacs was the backspace problem, which has largely been solved by now, at least in Linux. Later I learned to harness some of emacs's power, but I didn't find the initial learning curve a problem at all. vi was a different story, even with a cheat sheet.... So basically, working arrow keys and non-modal behavior are the most important requirements for a friendly editor learning curve. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From rfunk at funknet.net Sat Mar 6 00:56:07 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] Sound In-Reply-To: <20040306045723.GB5485@earthlink.net> References: <20040306045723.GB5485@earthlink.net> Message-ID: <200403060056.07823.rfunk@funknet.net> Steve Ferguson wrote: > I can't get any audio to work with XMMS (or anything else for that > matter). I've tried all combinations of > plugins, etc. Any ideas? Anyone work with this driver before? Did you check the mixer and make sure the volume is up and not muted? I can't tell whether you're using OSS or ALSA, but I believe ALSA defaults to muted. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From blitz at post891.org Sat Mar 6 05:23:55 2004 From: blitz at post891.org (Patrick Blitz) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] Sound In-Reply-To: <200403060056.07823.rfunk@funknet.net> References: <20040306045723.GB5485@earthlink.net> <200403060056.07823.rfunk@funknet.net> Message-ID: <1078568636.1238.2.camel@amon> On Sat, 2004-03-06 at 06:56, Rob Funk wrote: > Steve Ferguson wrote: > > I can't get any audio to work with XMMS (or anything else for that > > matter). I've tried all combinations of > > plugins, etc. Any ideas? Anyone work with this driver before? > > Did you check the mixer and make sure the volume is up and not muted? > > I can't tell whether you're using OSS or ALSA, but I believe ALSA defaults > to muted. it defenitivly does. i'm just try to get my two soundcards working (a AC'97 compliant cheap on-board one, and a Terratec Dmx6fire). try starting xmms from the command line, or just use mplayer or similar. U'll get a lot more output from it, and if it doesn't say anything about not beeing able to open the sound devices, than it will be the volume. kmix, aumix or even alsamixer will help you there. From jonadab at bright.net Sat Mar 6 07:22:28 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:40 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <200403060054.43518.rfunk@funknet.net> References: <4044D702.7020605@bluedogs-online.com> <20040304193900.4498e444.meffiem@neo.rr.com> <200403060054.43518.rfunk@funknet.net> Message-ID: Rob Funk writes: > Back when I started using email at OSU in 1992, I had a choice > between three different editors: emacs, vi, and one I'm pretty sure > was a home-brewed (at OSU) hack called fse. With no Unix experience > at the time, I chose emacs because it was the only one that seemed > to handle my arrow keys properly. [...] So basically, working > arrow keys and non-modal behavior are the most important > requirements for a friendly editor learning curve. This (arrow keys working properly) is the same reason I chose Emacs over vim when I starting messing with Debian. With Emacs, I could actually scroll through the documentation, if I carefully disciplined myself not to touch the home key. However, it was still a long way from friendly. Most other keys didn't work correctly. Home and end didn't work correctly, which effectively prevented me from using Emacs for real editing (until I figured out how to change key bindings). The arrow keys didn't work correctly in conjunction with ctrl and/or shift. And so on and so forth. I learned elisp and now have a comfortable Emacs-based editor environment, but in order to achieve that there's 600K of custom lisp *plus* a number of third-party modules installed that don't come with Emacs. Because, I love Emacs, but the defaults *suck*. I still can't be productive in an uncustomized Emacs. Now, some of that 600K wouldn't be necessary for users with lower requirements. For example, here's a function some people wouldn't need: (defun ued-mode-home () "Move to first non-whitespace character on line, but if already there go to column zero instead." (interactive) (let ((posn (point))) (setq mark-active nil) ; Remove this if you don't use pc-select (beginning-of-line-text) (if (= posn (point)) (beginning-of-line)))) This is the way I require the home key to behave, because that's how it behaved in UED (the editor that I used previously), and I became quite thoroughly addicted to this behavior. (It saves lots of keystrokes over the alternate approach of hitting home first and then using Ctrl-right or similar to position the cursor where you really want it.) But people who settle for Notepad don't need this (or don't know they need it); they'd be happy if home just went to the beginning of the line, which they can get by turning on pc-select IIRC (or maybe it was pc-keys or pc-bindings). Actually, most of that 600K is stuff like that, stuff a lot of users wouldn't need. Still, an out-of-the-box Emacs isn't suitable. Then again, as I noted, true end users don't use a standalone text editor anyway, so Emacs is perhaps not the best example. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Sat Mar 6 07:31:42 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Article on Linux as Multimedia PC In-Reply-To: <40469DFD.5090509@acm.org> References: <200402292351.58732.rfunk@funknet.net> <200403021155.16033.rfunk@funknet.net> <3147.209.239.137.161.1078248513.squirrel@209.239.137.161> <200403021247.10753.rfunk@funknet.net> <1078362220.2373.3.camel@linux.local> <40469DFD.5090509@acm.org> Message-ID: <8yief975.fsf@jonadab.homeip.net> Steve Roggenkamp writes: > Check out www.linuxbios.org. They claim a THREE second boot time as > their fastest time. Not sure if that's to a console prompt or not > though. That's not the whole boot time. It's just the BIOS boot time -- the time until the kernel starts loading. (Three seconds is still quite good; a lot of BIOSes take fifteen seconds or more. Then again, if you take the twenty-second part of the boot process and optimize it to three seconds, you're only gaining seventeen seconds. That's not so much against the time it takes to start the kernel, start up all the services, start X, start a desktop environment, start your apps, ...) -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From skippy at skippy.net Sat Mar 6 09:37:20 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 Message-ID: <4049E220.9070608@skippy.net> I noticed last night that KDE 3.2 had popped into Debian Unstable. A quick 'apt-get update; apt-get -u dist-upgrade' had me downloading it. A few snags caught me. I had xlibmesa3 installed, which is a metapackage to support older applications that aren't compiled against the new GL library naming convention. Having that installed prevented me from upgrading x-window-system-core. I removed xlibmesa3, and everything upgraded properly. After the upgrade, I logged out eager to log back into KDE 3.2. After logging in, I enjoyed looking at the default background, with no desktop icons, no toolbars, no _nothing_. *grrrr* Okay, I'll CTRL+ALT+F1 back to a console to poke around. Nope! My console shows a blinking cursor in the upper left corner of the screen, and nothing else. No noticeable response from any keyboard input. At least it responded to CTRL+ALT+DEL. Reboot into single user mode. Poke around a bit. Install a few other packages, to see if that helps (kdebase, for example, which wasn't installed). No change in symptoms: blank window with background. Nothing on text console. I _finally_ discovered that kdm had logged me into Sawfish for some reason. I middle-clicked (or maybe it was a right-click) and got a menu. Hallelujia! Exit back to kdm, login specifically selecting KDE. KDE 3.2 -- alright! However, my text console is still just a blinking cursor in the upper left. What's up with that? Can anyone tell me what's wrong, and how to fix it? Thanks! Scott From meffiem at neo.rr.com Sat Mar 6 10:20:02 2004 From: meffiem at neo.rr.com (Michael Meffie) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: References: <4044D702.7020605@bluedogs-online.com> <20040304193900.4498e444.meffiem@neo.rr.com> <200403060054.43518.rfunk@funknet.net> Message-ID: <20040306102002.6a4a4c0e.meffiem@neo.rr.com> > Then again, as I noted, true end users don't use a standalone text ^^^^ > editor anyway, so Emacs is perhaps not the best example. You mean *most* users don't use a text editor, even notepad. I don't think of Rob as being untrue just becuase he uses emacs :) I think what is needed is a usability lab. It would be nice if OSDL could sponsor such an effort. From dmaxwell at columbus.rr.com Sat Mar 6 10:36:29 2004 From: dmaxwell at columbus.rr.com (Dave Maxwell) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <4049E220.9070608@skippy.net> References: <4049E220.9070608@skippy.net> Message-ID: <200403061036.29273.dmaxwell@columbus.rr.com> Today 10:32:48 am On Saturday 06 March 2004 09:37 am, Scott Merrill wrote: ? > > I _finally_ discovered that kdm had logged me into Sawfish for some > reason. ?I middle-clicked (or maybe it was a right-click) and got a > menu. ?Hallelujia! ?Exit back to kdm, login specifically selecting KDE. > ? KDE 3.2 -- alright! > > However, my text console is still just a blinking cursor in the upper > left. ?What's up with that? ?Can anyone tell me what's wrong, and how to > fix it? > > Thanks! > Scott Issues like this also hosed me up on making a Custom Knoppix and I also upgraded my Sid machine last night. ?The only way I could get things remotely sane was to do this: 1. Log in using Windowmaker or some other WM. 2. Rename or delete .kde and .kderc 3. Open a shell and run startkde 4. Go through the wizard and let KDE create a sane environment for itself. 5. Quit KDE (the kicker is running at this point.) 6. Quit Windowmaker (or whatever you used) Now you can log in to KDE using KDM like normal and have a grand glorious time putting your desktop and Kmail filters back in order......... ?Oh, and just moving .kde .kderc and going into KDE straight away didn't work for me. ? I had to force the issue by having a WM already running. I suppose you could selectively copy things back in from your old .kde directory. ?With my luck, I'd make the damn thing sick again. ?I suppose this is something else that could be added to ESR's rant. ?This process should have Just Worked. ?The most I should have put up with is font setting being off or funky background selection. ?The KDE people do generally excellent work but something like this happens everytime I go one release to another. Dave From skippy at skippy.net Sat Mar 6 12:23:54 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <200403061036.29273.dmaxwell@columbus.rr.com> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> Message-ID: <404A092A.8040608@skippy.net> Dave Maxwell wrote: > On Saturday 06 March 2004 09:37 am, Scott Merrill wrote: > >>I _finally_ discovered that kdm had logged me into Sawfish for some >>reason. I middle-clicked (or maybe it was a right-click) and got a >>menu. Hallelujia! Exit back to kdm, login specifically selecting KDE. >> KDE 3.2 -- alright! > Issues like this also hosed me up on making a Custom Knoppix and I also > upgraded my Sid machine last night. The only way I could get things remotely > sane was to do this: > > 1. Log in using Windowmaker or some other WM. > 2. Rename or delete .kde and .kderc > 3. Open a shell and run startkde > 4. Go through the wizard and let KDE create a sane environment for itself. > 5. Quit KDE (the kicker is running at this point.) > 6. Quit Windowmaker (or whatever you used) > > Now you can log in to KDE using KDM like normal and have a grand glorious time > putting your desktop and Kmail filters back in order......... Oh, and just > moving .kde .kderc and going into KDE straight away didn't work for me. I > had to force the issue by having a WM already running. I didn't have to do any of this. Once I selected the KDE session in kdm, my desktop showed up, icons and taskbar intact. I don't use many KDE apps, so I don't know what minor settings may have been "reset". But so far all my preferences seem to have been preserved. > I suppose you could selectively copy things back in from your old .kde > directory. With my luck, I'd make the damn thing sick again. I suppose this > is something else that could be added to ESR's rant. This process should > have Just Worked. The most I should have put up with is font setting being > off or funky background selection. The KDE people do generally excellent > work but something like this happens everytime I go one release to another. My issue is the text console (CTRL+ALT+F1) being non-functional. That's pretty serious. I'm not even sure where to begin looking for answers. From george at byrd.net Sat Mar 6 12:23:54 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <404A092A.8040608@skippy.net> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> <404A092A.8040608@skippy.net> Message-ID: <404A092A.3050806@byrd.net> Hi Scott, I may be going out on a limb, but when I installed kernel 2.6.2, I did a fresh compile and forgot to check off UnixTTYs or something like that. You didn't mention recompiling your kernel, so I don't know how they would be related, but the end result is that I had the exact same problem you did with the flashing cursor in the upper left hand corner. There were also a couple of related modules I had to install along with the UnixTTYs (or whatever they were). Even if you didn't recompile your kernel, perhaps somehow those modules got deleted or turned off. Probably not much help, but then again, maybe it is... Good luck! George :) Scott Merrill wrote: > Dave Maxwell wrote: > >> On Saturday 06 March 2004 09:37 am, Scott Merrill wrote: >> >>> I _finally_ discovered that kdm had logged me into Sawfish for some >>> reason. I middle-clicked (or maybe it was a right-click) and got a >>> menu. Hallelujia! Exit back to kdm, login specifically selecting KDE. >>> KDE 3.2 -- alright! >> > > >> Issues like this also hosed me up on making a Custom Knoppix and I >> also upgraded my Sid machine last night. The only way I could get >> things remotely sane was to do this: >> >> 1. Log in using Windowmaker or some other WM. >> 2. Rename or delete .kde and .kderc >> 3. Open a shell and run startkde >> 4. Go through the wizard and let KDE create a sane environment for >> itself. >> 5. Quit KDE (the kicker is running at this point.) >> 6. Quit Windowmaker (or whatever you used) >> >> Now you can log in to KDE using KDM like normal and have a grand >> glorious time putting your desktop and Kmail filters back in >> order......... Oh, and just moving .kde .kderc and going into KDE >> straight away didn't work for me. I had to force the issue by >> having a WM already running. > > > I didn't have to do any of this. Once I selected the KDE session in > kdm, my desktop showed up, icons and taskbar intact. I don't use many > KDE apps, so I don't know what minor settings may have been "reset". > But so far all my preferences seem to have been preserved. > >> I suppose you could selectively copy things back in from your old >> .kde directory. With my luck, I'd make the damn thing sick again. I >> suppose this is something else that could be added to ESR's rant. >> This process should have Just Worked. The most I should have put up >> with is font setting being off or funky background selection. The >> KDE people do generally excellent work but something like this >> happens everytime I go one release to another. > > > My issue is the text console (CTRL+ALT+F1) being non-functional. > That's pretty serious. I'm not even sure where to begin looking for > answers. > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > From skippy at skippy.net Sat Mar 6 12:41:22 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <404A092A.3050806@byrd.net> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> <404A092A.8040608@skippy.net> <404A092A.3050806@byrd.net> Message-ID: <404A0D42.8050808@skippy.net> George Byrd wrote: > Hi Scott, > > I may be going out on a limb, but when I installed kernel 2.6.2, I did a > fresh compile and forgot to check off UnixTTYs or something like that. > You didn't mention recompiling your kernel, so I don't know how they > would be related, but the end result is that I had the exact same > problem you did with the flashing cursor in the upper left hand corner. > There were also a couple of related modules I had to install along with > the UnixTTYs (or whatever they were). > > Even if you didn't recompile your kernel, perhaps somehow those modules > got deleted or turned off. I'm using 2.6.2. It was working fine with KDE 3.1.5 and XFree 4.2. Maybe I'll give kernel 2.6.3 a whirl, and see if that helps. Thanks for the suggestion! From george at byrd.net Sat Mar 6 13:56:21 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <404A0D42.8050808@skippy.net> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> <404A092A.8040608@skippy.net> <404A092A.3050806@byrd.net> <404A0D42.8050808@skippy.net> Message-ID: <404A1ED5.6040004@byrd.net> Hi Scott, Just checking my .config file for 2.6.2 and the settings were: Character Devices -> [check] Virtual Terminal & Support for console on Virtual Terminal Character Devices -> [check] Unix 98 PTYs File Systems -> Pseudo Filesystems -> [check] /dev/pts filesystem for Unix98 PTYs After I enabled those and re-compiled, I got my login prompts back. Good luck, let us know how it went! George :) Scott Merrill wrote: > George Byrd wrote: > >> Hi Scott, >> >> I may be going out on a limb, but when I installed kernel 2.6.2, I >> did a fresh compile and forgot to check off UnixTTYs or something >> like that. You didn't mention recompiling your kernel, so I don't >> know how they would be related, but the end result is that I had the >> exact same problem you did with the flashing cursor in the upper left >> hand corner. There were also a couple of related modules I had to >> install along with the UnixTTYs (or whatever they were). >> >> Even if you didn't recompile your kernel, perhaps somehow those >> modules got deleted or turned off. > > > I'm using 2.6.2. It was working fine with KDE 3.1.5 and XFree 4.2. > Maybe I'll give kernel 2.6.3 a whirl, and see if that helps. > > Thanks for the suggestion! > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > From gate at ilive4code.net Sat Mar 6 14:25:59 2004 From: gate at ilive4code.net (Greg Sidelinger) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] RH3WS Message-ID: <1078601158.4546.28.camel@sanitarium> I managed to take down my Redhat 3 Workstation load last Wednesday. I spent a few min that night trying to bring it up but I was having no luck so I just booted into one of my backup loads. I just finally got cause of my downtime. Heres what happened. After work I suspended the laptop to ram like I do all the time. A few hours later after I got home it did not resume from suspend, batteries were still good. So I had to manually crash it. Could not access it though the network because no interfaces would have been up. Well rebooting it just stops after detecting the pcmcia controller. Odd, so I tried one more time but removed the wireless card, same thing. So now I'm a bit worried and figured I might have some file system damage. Forced a file system scan on all my partitions which turned up nothing. It was getting late and I had other things to do so I figured I would just fix it when I had more spare time. So I started to work on it again today and after about 30 min of working on it I got the sucker to boot. I ended up booting off the rescue cd and changed /etc/inittab so that I would boot into level 1 by default. I could not remember the boot param for this. After that the system booted. so I figured I would track down what was causing the hang. My gut told me pcmcia since the yenta controller was the last thing to go across my screen before the lockup. I did a service pcmcia from level 1 which gave me some kind of mismatch error, sorry, did not write it down. I then figured I would just remove pcmcia from the run levels and reboot. System came up just fine. Ok did a service pcmcia start and everything came right up. re-enabled pcmcia in level 3 and 5 and then rebooted again. Everything came up just fine and the system was back to its original setup. The logs don't say much other than pcmcia was the last thing it tried to start after failing. Now I'm just wondering if any one has any idea's as to what happened. I did not have any problems in the two day period with susu or xp with my pcmcia devices so I don't think it was hardware related. I'm just trying to figure out what would have caused this problem so if anyone has any ideas of what could have caused this please let me know. Greg From lefevre.10 at osu.edu Sat Mar 6 20:16:10 2004 From: lefevre.10 at osu.edu (Steve Lefevre) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] mencoder expert? Message-ID: <404A77DA.2060009@osu.edu> Is there anyone on the list willing to count themself as an mencoder expert? I have a 'debian' box from a knoppix HD install. I've gotten it to mplayer to compile. I was able to rip a region-free DVD. However, ripping an encoded DVD is another matter. I get a segmentation fault when it starts. What have I done wrong? (BTW debian knows my DVD drive as /dev/cdrom1, not /dev/dvd) $ mencoder /dev/cdrom1 -oac pcm -ovc lavc ./movie.avi MEncoder 0.92-3.3.3 (C) 2000-2003 MPlayer Team CPU: Advanced Micro Devices Duron MG Morgan (Family: 6, Stepping: 1) Detected cache-line size is 64 bytes CPUflags: Type: 6 MMX: 1 MMX2: 1 3DNow: 1 3DNow2: 1 SSE: 1 SSE2: 0 Reading /root/.mplayer/codecs.conf: can't open '/root/.mplayer/codecs.conf': No such file or directory Reading /usr/local/etc/mplayer/codecs.conf: 50 audio & 136 video codecs File not found: 'frameno.avi' Reading config file /root/.mplayer/mencoder: No such file or directory font: can't open file: /root/.mplayer/font/font.desc font: can't open file: /usr/local/share/mplayer/font/font.desc success: format: 0 data: 0x0 - 0xCE050000 MPEG-PS file format detected. MPEG: No audio stream found -> no sound. VIDEO: MPEG2 720x480 (aspect 2) 29.97 fps 9800.0 kbps (1225.0 kbyte/s) [V] filefmt:2 fourcc:0x10000002 size:720x480 fps:29.97 ftime:=0.0334 Opening video filter: [expand=-1:-1:-1:-1:1] Expand: -1 x -1, -1 ; -1 (-1=autodetect) osd: 1 ========================================================================== Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough VDec: vo config request - 720 x 480 (preferred csp: Mpeg PES) Could not find matching colorspace - retrying with -vop scale... Opening video filter: [scale] The selected video_out device is incompatible with this codec. VDecoder init failed :( Opening video decoder: [libmpeg2] MPEG 1/2 Video decoder v2.0 libmpeg2: Using MMXEXT for IDCT transform libmpeg2: Using MMXEXT for motion compensation VDec: vo config request - 720 x 480 (preferred csp: Planar YV12) VDec: using Planar YV12 as output csp (no 0) Movie-Aspect is 1.33:1 - prescaling to correct movie aspect. videocodec: libavcodec (720x480 fourcc=58564944 [DIVX]) Selected video codec: [mpeg12] vfm:libmpeg2 (MPEG 1 or 2 (libmpeg2)) ========================================================================== Writing AVI header... Segmentation faultf ( 0%) 0fps Trem: 0min 0mb A-V:0.000 [0:0] From rcrews at nc.rr.com Sat Mar 6 09:15:50 2004 From: rcrews at nc.rr.com (Randy Crews) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Solaris Admin Message-ID: <00f601c40385$8ace3500$6601a8c0@PVADOM.pvaglobal.com> I'm not sure if I'm posting to the right mailing list but here goes. I have an immediate need for a Solaris Administrator in Dublin Ohio. Please send resume to resumes@pvaglobal.com to get more information. Thanks Randy Crews VP Network Services www.pvaglobal.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040306/469d2648/attachment.htm From skippy at skippy.net Sun Mar 7 14:44:14 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <404A1ED5.6040004@byrd.net> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> <404A092A.8040608@skippy.net> <404A092A.3050806@byrd.net> <404A0D42.8050808@skippy.net> <404A1ED5.6040004@byrd.net> Message-ID: <404B7B8E.708@skippy.net> George Byrd wrote: > Hi Scott, > > Just checking my .config file for 2.6.2 and the settings were: > > Character Devices -> [check] Virtual Terminal & Support for console on > Virtual Terminal > Character Devices -> [check] Unix 98 PTYs > File Systems -> Pseudo Filesystems -> [check] /dev/pts filesystem for > Unix98 PTYs > > After I enabled those and re-compiled, I got my login prompts back. I compiled and installed kernel 2.6.3, making extra sure to enable those options per George's suggestion. When in X, CTRL+ALT+F1 gives me a functional console. Thanks George! From jonadab at bright.net Sun Mar 7 14:46:02 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] To need a tutorial is a failure of the software In-Reply-To: <20040306102002.6a4a4c0e.meffiem@neo.rr.com> References: <4044D702.7020605@bluedogs-online.com> <20040304193900.4498e444.meffiem@neo.rr.com> <200403060054.43518.rfunk@funknet.net> <20040306102002.6a4a4c0e.meffiem@neo.rr.com> Message-ID: <1xo4fnk5.fsf@jonadab.homeip.net> Michael Meffie writes: > You mean *most* users don't use a text editor, even notepad. I don't > think of Rob as being untrue just becuase he uses emacs :) It is his end-userness, not himself, that I implied is untrue. > I think what is needed is a usability lab. It would be nice if OSDL > could sponsor such an effort. A usability lab would be helpful, but this sort of thing can also be done in a distributed manner. What is needed is for powerusers (or developers, but there are more poweruser-man-hours available) to spend time watching end users use the software, taking notes on their usage patterns and on where they get hung up, discussing it with them, and so on -- and then communicate findings to developers and/or QA people. And, again, I don't think that features or UI elements that are found to be troublesome for end users should be removed from the software; rather, they should be made optional and turned off by default, so that powerusers can still choose to enable them, but end users don't have to be bothered by them. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From george at byrd.net Sun Mar 7 17:45:43 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <404B7B8E.708@skippy.net> References: <4049E220.9070608@skippy.net> <200403061036.29273.dmaxwell@columbus.rr.com> <404A092A.8040608@skippy.net> <404A092A.3050806@byrd.net> <404A0D42.8050808@skippy.net> <404A1ED5.6040004@byrd.net> <404B7B8E.708@skippy.net> Message-ID: <404BA617.8000406@byrd.net> Hooray! Glad I could be of help... George Scott Merrill wrote: > George Byrd wrote: > >> Hi Scott, >> >> Just checking my .config file for 2.6.2 and the settings were: >> >> Character Devices -> [check] Virtual Terminal & Support for console >> on Virtual Terminal >> Character Devices -> [check] Unix 98 PTYs >> File Systems -> Pseudo Filesystems -> [check] /dev/pts filesystem for >> Unix98 PTYs >> >> After I enabled those and re-compiled, I got my login prompts back. > > > I compiled and installed kernel 2.6.3, making extra sure to enable > those options per George's suggestion. > > When in X, CTRL+ALT+F1 gives me a functional console. > > Thanks George! > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > From bnmille at myrealbox.com Sun Mar 7 21:39:20 2004 From: bnmille at myrealbox.com (Brian Miller) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Re: Debian Unstable: KDE 3.2 and XFree 4.3 In-Reply-To: <200403071700.i27H0cs3031072@stones.wcbe.org> References: <200403071700.i27H0cs3031072@stones.wcbe.org> Message-ID: <200403072139.20734.bnmille@myrealbox.com> On Sunday 07 March 2004 12:00 pm, colug-request@colug.net wrote: > Date: Sat, 06 Mar 2004 12:23:54 -0500 > From: Scott Merrill >My issue is the text console (CTRL+ALT+F1) being >non-functional. ?That's pretty serious. ?I'm not even sure where to >begin looking for answers. This seems pretty obvious, but have you checked /etc/inittab to see if the text consoles are actually enabled? You should have some lines that look like this: # Run gettys in standard runlevels 1:2345:respawn:/sbin/mingetty tty1 2:2345:respawn:/sbin/mingetty tty2 #3:2345:respawn:/sbin/mingetty tty3 #4:2345:respawn:/sbin/mingetty tty4 #5:2345:respawn:/sbin/mingetty tty5 #6:2345:respawn:/sbin/mingetty tty6 In my case, I've disabled consoles 3-6, since I don't use them. From JBalint at alldata.net Mon Mar 8 08:51:27 2004 From: JBalint at alldata.net (Balint, Jess) Date: Sat Jan 8 01:34:41 2005 Subject: [COLUG] Freenode IRC Message-ID: <5789A42DD684734298BF9EC852D35111033AB28D@scad1exis02.alldata.net> Is anybody getting irc.freenode.net resolving to 127.0.0.1? I thought the admins here were being pissy, but I tried online nslookup and 127.0.0.1 also....? ~ From skippy at skippy.net Mon Mar 8 09:02:08 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB28D@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB28D@scad1exis02.alldata.net> Message-ID: <32923.216.136.35.122.1078754528.squirrel@www.skippy.net> Balint, Jess said: > Is anybody getting irc.freenode.net resolving to 127.0.0.1? I thought the > admins here were being pissy, but I tried online nslookup and 127.0.0.1 > also....? >From two different systems on seperate networks, using different DNS servers: skippy@server1:~$ host irc.freenode.net irc.freenode.net has address 213.92.8.4 irc.freenode.net has address 66.54.82.254 irc.freenode.net has address 69.28.212.219 irc.freenode.net has address 81.209.176.2 irc.freenode.net has address 82.96.64.2 irc.freenode.net has address 128.193.0.29 irc.freenode.net has address 128.193.0.47 irc.freenode.net has address 212.204.214.114 irc.freenode.net has address 213.28.116.205 [smerrill@smerrill smerrill]$ host irc.freenode.net irc.freenode.net has address 213.28.116.205 irc.freenode.net has address 213.92.8.4 irc.freenode.net has address 66.54.82.254 irc.freenode.net has address 69.28.212.219 irc.freenode.net has address 81.209.176.2 irc.freenode.net has address 82.96.64.2 irc.freenode.net has address 128.193.0.29 irc.freenode.net has address 128.193.0.47 irc.freenode.net has address 212.204.214.114 From jep200305 at columbus.rr.com Mon Mar 8 09:06:15 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB28D@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB28D@scad1exis02.alldata.net> Message-ID: <20040308090615.3d657020.jep200305@columbus.rr.com> On Mon, 08 Mar 2004 08:51:27 -0500 "Balint, Jess" wrote: > Is anybody getting irc.freenode.net resolving to 127.0.0.1? I thought the > admins here were being pissy, but I tried online nslookup and 127.0.0.1 > also....? [iwk@testhat iwk]$ host irc.freenode.net irc.freenode.net has address 81.209.176.2 irc.freenode.net has address 82.96.64.2 irc.freenode.net has address 128.193.0.29 irc.freenode.net has address 128.193.0.47 irc.freenode.net has address 212.204.214.114 irc.freenode.net has address 213.28.116.205 irc.freenode.net has address 213.92.8.4 irc.freenode.net has address 66.54.82.254 irc.freenode.net has address 69.28.212.219 [iwk@testhat iwk]$ nslookup irc.freenode.net Note: nslookup is deprecated and may be removed from future releases. Consider using the `dig' or `host' programs instead. Run nslookup with the `-sil[ent]' option to prevent this message from appearing. Server: 24.95.80.41 Address: 24.95.80.41#53 Non-authoritative answer: Name: irc.freenode.net Address: 128.193.0.29 Name: irc.freenode.net Address: 128.193.0.47 Name: irc.freenode.net Address: 212.204.214.114 Name: irc.freenode.net Address: 213.28.116.205 Name: irc.freenode.net Address: 213.92.8.4 Name: irc.freenode.net Address: 66.54.82.254 Name: irc.freenode.net Address: 69.28.212.219 Name: irc.freenode.net Address: 81.209.176.2 Name: irc.freenode.net Address: 82.96.64.2 [iwk@testhat iwk]$ From JBalint at alldata.net Mon Mar 8 09:13:32 2004 From: JBalint at alldata.net (Balint, Jess) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC Message-ID: <5789A42DD684734298BF9EC852D35111033AB28E@scad1exis02.alldata.net> Hrmm.. ok. Well it seems to be working now. Thanks guys. > -----Original Message----- > From: colug-bounces@colug.net > [mailto:colug-bounces@colug.net]On Behalf > Of Balint, Jess > Sent: Monday, March 08, 2004 8:51 AM > To: 'colug@colug.net' > Subject: [COLUG] Freenode IRC > > > Is anybody getting irc.freenode.net resolving to 127.0.0.1? I > thought the > admins here were being pissy, but I tried online nslookup and > 127.0.0.1 > also....? > > ~ From pat at linuxcolumbus.com Mon Mar 8 09:43:34 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB28E@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB28E@scad1exis02.alldata.net> Message-ID: <20040308144334.GA14427@linuxcolumbus.com> Works here too. Lurking on #chmlug from time to time. Pat On Mon, Mar 08, 2004 at 09:13:32AM -0500, Balint, Jess wrote: > Hrmm.. ok. Well it seems to be working now. Thanks guys. > > > -----Original Message----- > > From: colug-bounces@colug.net > > [mailto:colug-bounces@colug.net]On Behalf > > Of Balint, Jess > > Sent: Monday, March 08, 2004 8:51 AM > > To: 'colug@colug.net' > > Subject: [COLUG] Freenode IRC > > > > > > Is anybody getting irc.freenode.net resolving to 127.0.0.1? I > > thought the > > admins here were being pissy, but I tried online nslookup and > > 127.0.0.1 > > also....? > > > > ~ > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > From JBalint at alldata.net Mon Mar 8 10:45:10 2004 From: JBalint at alldata.net (Balint, Jess) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC Message-ID: <5789A42DD684734298BF9EC852D35111033AB290@scad1exis02.alldata.net> I think you mean #cmhlug? > -----Original Message----- > From: colug-bounces@colug.net > [mailto:colug-bounces@colug.net]On Behalf > Of pat@linuxcolumbus.com > Sent: Monday, March 08, 2004 9:44 AM > To: Central OH Linux User Group > Subject: Re: [COLUG] Freenode IRC > > > Works here too. > > Lurking on #chmlug from time to time. > > Pat > > > On Mon, Mar 08, 2004 at 09:13:32AM -0500, Balint, Jess wrote: > > Hrmm.. ok. Well it seems to be working now. Thanks guys. > > > > > -----Original Message----- > > > From: colug-bounces@colug.net > > > [mailto:colug-bounces@colug.net]On Behalf > > > Of Balint, Jess > > > Sent: Monday, March 08, 2004 8:51 AM > > > To: 'colug@colug.net' > > > Subject: [COLUG] Freenode IRC > > > > > > > > > Is anybody getting irc.freenode.net resolving to 127.0.0.1? I > > > thought the > > > admins here were being pissy, but I tried online nslookup and > > > 127.0.0.1 > > > also....? > > > > > > ~ > > _______________________________________________ > > colug mailing list > > colug@colug.net > > http://www.colug.net/mailman/listinfo/colug > > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > From pat at linuxcolumbus.com Mon Mar 8 10:53:27 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Freenode IRC In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB290@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB290@scad1exis02.alldata.net> Message-ID: <20040308155327.GC14427@linuxcolumbus.com> On Mon, Mar 08, 2004 at 10:45:10AM -0500, Balint, Jess wrote: > I think you mean #cmhlug? > Yes Sir. Pat From beano21 at direcway.com Mon Mar 8 16:36:09 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Gentoo Install In-Reply-To: <20040308155327.GC14427@linuxcolumbus.com> Message-ID: What a record. That was the longest install I have ever done in my life. And there is still more software to load. I followed the instructions, and since this is the second try I thought it would be easier. Now one of two things happened I either tried a different install type or the instructions changed. I stayed up until I had a running server.(maybe I did get two hours while it was running) Completed everything needed for a running system, then only the bootloader was left. Created lilo.conf, and rebooted. OH crap I forgot to run lilo. To late it has already shutdown.. I am taking a break for now. What would be the fastest way to log into the new Gentoo Install and configure and set lilo(server offsite)? I did manage to startup with the disc and mount all of the drives and then tried to shroot into it, but for some reason it didn't work. Any help on that would be great. On the install side I have never had so many options before in my life. Which Kernel to use(gs-sources), what to compile into the kernel or make modules. I am very pleased. Besides it is the first time I know exactly what is and isn't installed, and how they all fit together. I still consider myself a newbie and think this was be best thing for me. I enjoyed the instructions because they not only said what to do but in most cases it explained why and or linked to where you can get more information. Just thought I would throw out this email. Most of the people I know whold not understand nor appreciate any of it. BTW: I cannot use knoppix to use as a rescue disc. For some odd reason the cd-rom unit on this server is a tad shifty, and only likes certain disc's. _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From blata at extent0006.entomology.ohio-state.edu Mon Mar 8 16:54:05 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] nvidia modules Message-ID: <404CEB7D.8050809@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am trying to sooped my video card. The first step to try to get the card to use nvagp driver. However, my system keeps using agpgart. How can I get my system to stop loading agpgart. Nvidia readme says to add the "option nvagp=1" to the XF86Config file to load the alternate driver. But that wont happen as long as agpgart keeps loading. Im running RH 9. If anybody knows how to get FastWrites and Sideband Addressing to work I would appreciate the info. :) - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFATOt9v+6+qSFv3TURAho+AKC2nUOajVkewezI0+ZJerk4K7gR8wCgkFJA DYWwEHtlOZkRfx/25cCugAI= =/Nbn -----END PGP SIGNATURE----- From joe at whipple.cc Mon Mar 8 17:26:11 2004 From: joe at whipple.cc (Joe Whipple) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Gentoo Install In-Reply-To: References: Message-ID: <404CF303.90204@whipple.cc> The Gentoo CD is the perfect Gentoo Rescue disk.. Boot the CD mount /dev/hdxx /mnt/gentoo mount /dev/hdxx /mnt/gentoo/boot mount -t proc proc /mnt/gentoo/proc chroot /mnt/gentoo /bin/bash run lilo :) Frank Rieder wrote: >What a record. That was the longest install I have ever done in my life. >And there is still more software to load. I followed the instructions, and >since this is the second try I thought it would be easier. Now one of two >things happened I either tried a different install type or the instructions >changed. I stayed up until I had a running server.(maybe I did get two >hours while it was running) Completed everything needed for a running >system, then only the bootloader was left. Created lilo.conf, and rebooted. >OH crap I forgot to run lilo. To late it has already shutdown.. I am >taking a break for now. What would be the fastest way to log into the new >Gentoo Install and configure and set lilo(server offsite)? I did manage to >startup with the disc and mount all of the drives and then tried to shroot >into it, but for some reason it didn't work. Any help on that would be >great. > >On the install side I have never had so many options before in my life. >Which Kernel to use(gs-sources), what to compile into the kernel or make >modules. I am very pleased. Besides it is the first time I know exactly >what is and isn't installed, and how they all fit together. I still >consider myself a newbie and think this was be best thing for me. I enjoyed >the instructions because they not only said what to do but in most cases it >explained why and or linked to where you can get more information. > >Just thought I would throw out this email. Most of the people I know whold >not understand nor appreciate any of it. > > >BTW: I cannot use knoppix to use as a rescue disc. For some odd reason the >cd-rom unit on this server is a tad shifty, and only likes certain disc's. > > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug >--- >Incoming mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > >--- >Outgoing mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > > From beano21 at direcway.com Mon Mar 8 18:54:29 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] Gentoo Install In-Reply-To: <404CF303.90204@whipple.cc> Message-ID: Thanks Joe, I did try that but I did not mount proc. Thats prob why huh. And I tell ya the weirdest /dev for my harddrives I have ever seen. I have a Compaq Proliant the raid card handles the harddrives and it has them listed as: /dev/ida/disc0/part1-5 /dev/ida/disc1/part1-2 Maybe because I have only used ATA not SCSI or RAID arrays for that much. I knew it hade to be easy but I just got tired of messing with it for now. Maybe I will try tomarrow. -----Original Message----- From: colug-bounces@colug.net [mailto:colug-bounces@colug.net]On Behalf Of Joe Whipple Sent: Monday, March 08, 2004 5:26 PM To: Central OH Linux User Group Subject: Re: [COLUG] Gentoo Install The Gentoo CD is the perfect Gentoo Rescue disk.. Boot the CD mount /dev/hdxx /mnt/gentoo mount /dev/hdxx /mnt/gentoo/boot mount -t proc proc /mnt/gentoo/proc chroot /mnt/gentoo /bin/bash run lilo :) Frank Rieder wrote: >What a record. That was the longest install I have ever done in my life. >And there is still more software to load. I followed the instructions, and >since this is the second try I thought it would be easier. Now one of two >things happened I either tried a different install type or the instructions >changed. I stayed up until I had a running server.(maybe I did get two >hours while it was running) Completed everything needed for a running >system, then only the bootloader was left. Created lilo.conf, and rebooted. >OH crap I forgot to run lilo. To late it has already shutdown.. I am >taking a break for now. What would be the fastest way to log into the new >Gentoo Install and configure and set lilo(server offsite)? I did manage to >startup with the disc and mount all of the drives and then tried to shroot >into it, but for some reason it didn't work. Any help on that would be >great. > >On the install side I have never had so many options before in my life. >Which Kernel to use(gs-sources), what to compile into the kernel or make >modules. I am very pleased. Besides it is the first time I know exactly >what is and isn't installed, and how they all fit together. I still >consider myself a newbie and think this was be best thing for me. I enjoyed >the instructions because they not only said what to do but in most cases it >explained why and or linked to where you can get more information. > >Just thought I would throw out this email. Most of the people I know whold >not understand nor appreciate any of it. > > >BTW: I cannot use knoppix to use as a rescue disc. For some odd reason the >cd-rom unit on this server is a tad shifty, and only likes certain disc's. > > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug >--- >Incoming mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > >--- >Outgoing mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > > _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From dmaxwell at columbus.rr.com Mon Mar 8 22:41:38 2004 From: dmaxwell at columbus.rr.com (Dave Maxwell) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] nvidia modules In-Reply-To: <404CEB7D.8050809@bugs.osu.edu> References: <404CEB7D.8050809@bugs.osu.edu> Message-ID: <200403082241.38739.dmaxwell@columbus.rr.com> On Monday 08 March 2004 04:54 pm, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I am trying to sooped my video card. The first step to try to get the > card to use nvagp driver. However, my system keeps using agpgart. > How can I get my system to stop loading agpgart. Nvidia readme says agpgart is being loaded in modules, modules.conf or by modconf. There isn't anything you feed to the NV drivers to prevent that. You have to tweak your module config not to load agpgart. > If anybody knows how to get FastWrites and Sideband Addressing to work > I would appreciate the info. :) I found this on a forum site: To enable fast writes and SBA, simply add the support variables that your card supports. JUST after 'alias char-major-195 nvidia' in /etc/modules.conf add one (or both if supported) add this line. options nvidia NVreg_EnableAGPSBA=1 NVreg_EnableAGPFW=1 I'd be double dog sure your particular card supports those before I do it. You may also have to enable Fast Write in you machine's bios. Dave -- The real value of KDE is that they inspired and push the development of GNOME :-) -- #Debian From blata at extent0006.entomology.ohio-state.edu Mon Mar 8 23:04:41 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] nvidia modules In-Reply-To: <200403082241.38739.dmaxwell@columbus.rr.com> References: <404CEB7D.8050809@bugs.osu.edu> <200403082241.38739.dmaxwell@columbus.rr.com> Message-ID: <404D4259.6090705@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 thanks for the info. modules.conf was my guess too but there is no mention of agpgart in it. alias eth0 8139too alias eth1 natsemi alias usb-controller ehci-hcd alias usb-controller1 usb-uhci alias sound-slot-0 es1371 post-install sound-slot-0 /bin/aumix-minimal -f /etc/.aumixrc -L |/dev/null 2>&1 || : pre-remove sound-slot-0 /bin/aumix-minimal -f /etc/.aumixrc -S |/dev/null 2>&1 || : alias char-major-195 nvidia So where is it written to load the agpgart module? I have checked /proc/drivers/nvidia/card and it is capable. Its a Ti4400. You wrote this to me on 03/08/2004 10:41 PM: |On Monday 08 March 2004 04:54 pm, Wade Pinkston wrote: | |>-----BEGIN PGP SIGNED MESSAGE----- |>Hash: SHA1 |> |>I am trying to sooped my video card. The first step to try to get the |>card to use nvagp driver. However, my system keeps using agpgart. |>How can I get my system to stop loading agpgart. Nvidia readme says | | |agpgart is being loaded in modules, modules.conf or by modconf. There isn't |anything you feed to the NV drivers to prevent that. You have to tweak your |module config not to load agpgart. | |>If anybody knows how to get FastWrites and Sideband Addressing to work |>I would appreciate the info. :) | | | |I found this on a forum site: | |To enable fast writes and SBA, simply add the support variables that your card |supports. JUST after |'alias char-major-195 nvidia' in /etc/modules.conf add one (or both if |supported) add this line. |options nvidia NVreg_EnableAGPSBA=1 NVreg_EnableAGPFW=1 | | |I'd be double dog sure your particular card supports those before I do it. |You may also have to enable Fast Write in you machine's bios. | |Dave | - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFATUJZv+6+qSFv3TURAsbfAJ9w97+mB1cugtPaSO4g8cpmMZFYTACePyuK pbhkqKTkVViNx0b8CNl45ao= =EHsz -----END PGP SIGNATURE----- From cacepi at adelphia.net Mon Mar 8 23:59:46 2004 From: cacepi at adelphia.net (Jason Martens) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] nvidia modules In-Reply-To: <404D4259.6090705@bugs.osu.edu> References: <404CEB7D.8050809@bugs.osu.edu> <200403082241.38739.dmaxwell@columbus.rr.com> <404D4259.6090705@bugs.osu.edu> Message-ID: <404D4F42.5070404@adelphia.net> Wade Pinkston wrote: > thanks for the info. modules.conf was my guess too but there is no > mention of agpgart in it. > > So where is it written to load the agpgart module? If depends on the distro, but I'd say the culprit is a framebuffer console (high color), which usually implies loading the agpgart driver. Check your boot loader and see if there is a vga option stated ("vga=791", for example). Jason Martens cacepi@adelphia.net From bruceobe at fastmail.fm Tue Mar 9 11:05:15 2004 From: bruceobe at fastmail.fm (Bruce Obenour) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] question about swap list or for sale list Message-ID: <404DEB3B.9020601@fastmail.fm> All, Does anyone know of a good mailing list or forum where I can post a list of hardware (intel) and monitors I need to unload ? I'm selling cheap or giving away things. I am a member of a MAC list, but I know of no Intel list. Ideas ? Many thanks Bruce From beano21 at direcway.com Tue Mar 9 18:29:37 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] question about swap list or for sale list In-Reply-To: <404DEB3B.9020601@fastmail.fm> Message-ID: I don't know of a list per se but I have been looking for some mac stuff. Can you provide me the link for the Mac List you mentioned? Messed up a mac trying to get Linux on it.lol. -----Original Message----- From: colug-bounces@colug.net [mailto:colug-bounces@colug.net]On Behalf Of Bruce Obenour Sent: Tuesday, March 09, 2004 11:05 AM To: Central OH Linux User Group Subject: [COLUG] question about swap list or for sale list All, Does anyone know of a good mailing list or forum where I can post a list of hardware (intel) and monitors I need to unload ? I'm selling cheap or giving away things. I am a member of a MAC list, but I know of no Intel list. Ideas ? Many thanks Bruce _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From blata at extent0006.entomology.ohio-state.edu Tue Mar 9 18:35:10 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:42 2005 Subject: [COLUG] nvidia modules In-Reply-To: <404D4F42.5070404@adelphia.net> References: <404CEB7D.8050809@bugs.osu.edu> <200403082241.38739.dmaxwell@columbus.rr.com> <404D4259.6090705@bugs.osu.edu> <404D4F42.5070404@adelphia.net> Message-ID: <404E54AE.6080508@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I got fast writes and sba turned on thanks to this site. http://www.brandonhutchinson.com/NVIDIA_drivers_with_Red_Hat_Linux.html You wrote this to me on 03/08/2004 11:59 PM: | Wade Pinkston wrote: | |> thanks for the info. modules.conf was my guess too but there is no |> mention of agpgart in it. |> |> So where is it written to load the agpgart module? | | | If depends on the distro, but I'd say the culprit is a | framebuffer console (high color), which usually implies loading the | agpgart driver. Check your boot loader and see if there is a vga | option stated ("vga=791", for example). | | Jason Martens | cacepi@adelphia.net | | _______________________________________________ | colug mailing list | colug@colug.net | http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFATlSuv+6+qSFv3TURAgECAJ4p003nVHxW1B+uaraAMVuYPnt84QCeKyXb SpciLkfol49QUIY74HGRrwA= =cGOb -----END PGP SIGNATURE----- From tom at functionalmedia.com Tue Mar 9 19:17:09 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] Zope useage, CMF PLONE Message-ID: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> Colug, Is anyone in Colug using ZOPE ?? I am reviewing it for potential use on a new project. I am unsure about how I feel about it. Seems like a great idea but I have my old habits of doing everyhting by hand with php/mysql. Zope seems like a step outside of that old pattern. I need a site that gathers dynamic content and displays it. I need to have regular folks add content. I may access RSS feeds. My path is clear with php/mysql. Put the data in mysql, write some hasty php scripts to access the data, write some password protected pages to allow administration and I am done. With ZOPE it looks like it takes more planning at the start. Build the structure. Write the methods. Add users and define roles. Build interfaces for administration. When I look at ZOPE and feel tempted to write an application without using the ZOPE features it seems like I should not bother with ZOPE at all. Meaning if I do not allow the use of the resource/url/manage function than why bother? Anyone building in ZOPE ?? ANyone using PLONE or the ZOPE CMF?? Please advise... Good book would be helpful. DTML vs the TAL templetaing language for example. Regards, Tom Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From anderson at wks.uts.ohio-state.edu Tue Mar 9 20:05:37 2004 From: anderson at wks.uts.ohio-state.edu (Bill Anderson) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] Zope useage, CMF PLONE In-Reply-To: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> Message-ID: > Is anyone in Colug using ZOPE ?? I've been developing a Zope/Plone site for the past several months. Overall I like it. It is definitely worthwhile if you have a need for allowing multiple people to author/edit/approve website content or if you need more granular access control than basic auth provides. I have found the documentation to be somewhat lacking and the learning curve to be rather steep, but if I was to do it over again with a chance to choose another language/system, I'd probably still use Zope/Plone. > I am reviewing it for potential use on a new project. I am unsure about > how I feel about it. Seems like a great idea but I have my old habits > of doing everyhting by hand with php/mysql. Zope seems like a step > outside of that old pattern. Takes some getting used to, but worthwhile IMHO. > I need a site that gathers dynamic content and displays it. I need to > have regular folks add content. I may access RSS feeds. >From what I understand, Zope has good RSS support, although I havn't used it yet. > With ZOPE it looks like it takes more planning at the start. Build the > structure. Write the methods. Add users and define roles. Build > interfaces for administration. Yup. Nicer than SQL calls though. > When I look at ZOPE and feel tempted to write an application without > using the ZOPE features it seems like I should not bother with ZOPE at > all. Meaning if I do not allow the use of the resource/url/manage > function than why bother? I think the results with Zope are much cleaner and have better maintainability. I'd choose Zope if you have adequate development time to get past the learning curve and plan to maintain it for a while. Also good if you want to have your code reviewed/audited because it is easier to keep presentable. Bill From rutledge.50 at osu.edu Wed Mar 10 09:35:19 2004 From: rutledge.50 at osu.edu (LINCOLN RUTLEDGE) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] NIC problem... Message-ID: <82993782c8da.82c8da829937@osu.edu> Good morning, Okay, I was hoping for any insight into a problem I'm seeing. I have about 50 machines with on-board 3com 10/100 LAN. A few of them are completely losing network connectivity. No ping, no nothing. The lights look normal. Unplug the LAN cable, plug it back in, voila. Back up and running. Usually. I swapped one person computers, and they are still having the problem, though not as often. I saw the problem on their machine in my office once, and I remember it happening on my old machine in my office a while back. Only under Windows, I never had any problem under any other OS that I remember. We were having the problem before we upgraded to a switched backbone. It seems to be better now for some people, but it is still happening daily for one specific user (the one I switched with). I have replaced everything in their office, computer, cables, clean image of the OS. I replaced the line going from the patch bay to the switch, and I tested their cat-5 run to make sure it's wired right and has continuity. I think it's these on-board LANs. Any ideas? They are stupid micro-whatever form factor machines, so I can't use normal pci cards in them (grrr...) I am going to see if they have flash firmware and try some drivers. I also tried forcing them to 10 MB or 100MB half-duplex, but it did nothing. I guess I will try a laptop in that person's office with a real 3Com PCMCIA nic and see how it works... Thanks, Lincoln From pat at linuxcolumbus.com Wed Mar 10 10:00:25 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] NIC problem... In-Reply-To: <82993782c8da.82c8da829937@osu.edu> References: <82993782c8da.82c8da829937@osu.edu> Message-ID: <20040310150025.GA17638@linuxcolumbus.com> On Wed, Mar 10, 2004 at 09:35:19AM -0500, LINCOLN RUTLEDGE wrote: > I think it's these on-board LANs. Any ideas? They are stupid micro-whatever form factor machines, so I can't use normal pci cards in them (grrr...) > Sounds like the on board nics are bee-roke. Just because they are 3com's doesn't make them any better than the cheap $5 nics from MC. I've got one that will take down an unswitched network it is so chatty. Pat From lefevre.10 at osu.edu Wed Mar 10 19:21:23 2004 From: lefevre.10 at osu.edu (STEVEN LEFEVRE) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] date/time libraries in python, and general date/time questions Message-ID: <8dd5108dd0a0.8dd0a08dd510@osu.edu> Hey folks - I'm currently learning python. I just found out that python has no native date types or functions. Could any python hacker refer me to a library they particularly like? I'm writing a system (for lack of a better term) that will convert English-like date phrases (such as 'the day before yesterdays' or 'five days after the first Tuesday in next November'), so I'll be doing a lot of adding and subtracting, but also I will be needed to check the name of the month and day in the date variable. I'm not sure whether to manipulate unix timestamps, or use a more conventional date format such as DD-MM-YYYY. Steve Lefevre From drlinux at columbus.rr.com Wed Mar 10 19:43:52 2004 From: drlinux at columbus.rr.com (Dave Reed) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] Re: date/time libraries in python, and general date/time questions In-Reply-To: <8dd5108dd0a0.8dd0a08dd510@osu.edu> References: <8dd5108dd0a0.8dd0a08dd510@osu.edu> Message-ID: <200403101943.52385.drlinux@columbus.rr.com> On Wednesday 10 March 2004 19:21, STEVEN LEFEVRE wrote: > Hey folks - > > I'm currently learning python. I just found out that python has no native date types or functions. > > Could any python hacker refer me to a library they particularly like? I'm writing a system (for lack of a better term) that will convert English-like date phrases (such as 'the day before yesterdays' or 'five days after the first Tuesday in next November'), so I'll be doing a lot of adding and subtracting, but also I will be needed to check the name of the month and day in the date variable. > > I'm not sure whether to manipulate unix timestamps, or use a more conventional date format such as DD-MM-YYYY. > > Steve Lefevre This is the one most people use. http://www.egenix.com/files/python/mxDateTime.html I can't remember if it's now included with Python 2.3 or if you still have to install it separately yourself. Or it's possible your distribution may include it even if it doesn't come with Python. Looks like it comes with Red Hat 9 (I think the PostgreSQL Python bindings require it which Red Hat includes): rpm -qf /usr/lib/python2.2/site-packages/mx/DateTime/mxDateTime mx-2.0.3-8 but is not part of Python 2.2. If this line works in the Python interpreter, you already have it: from mx.DateTime import DateTime If not, download and install it. Dave From sun at percipia.com Thu Mar 11 18:44:50 2004 From: sun at percipia.com (Sundaram Ramasamy) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> Message-ID: <002101c407c2$e01e2ba0$6400a8c0@sundaram> Hi, I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do ls or commenting the shell, it displays in blue color, it was visible. How will I change the color? disable the color. -SR From jep200305 at columbus.rr.com Thu Mar 11 19:19:25 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <002101c407c2$e01e2ba0$6400a8c0@sundaram> References: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> <002101c407c2$e01e2ba0$6400a8c0@sundaram> Message-ID: <20040311191925.3e070bf6.jep200305@columbus.rr.com> Sundaram Ramasamy wrote: > I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do > ls or commenting the shell, it displays in blue color, it was visible. > > How will I change the color? disable the color. This will likely take a variety of changes in a variety of places. Maybe you can tell putty to tell the remote computer, that your (putty) terminal is not capable of displaying color. You would find an old or contrive a new colorless terminal in termcap on the remote computer and then tell putty to tell the remote computer that it is that colorless terminal. Hacking termcap entries (or its replacement) can be intimidating, but just removing colors should not be too hard. That would be a rather global change. Other changes would be to individually configure each program, such as ls, perhaps with config files, or alias tricks such as alias ls='ls --color=never' in ~/.bashrc, /etc/bashrc or /etc/skel/.bashrc, depending on for how many users you want to make this change for. ~/.bashrc is probably good enough for you. You'll probably want to read the dox for vim and disable colors or the syntax highlighting that provokes them in ~/.vimrc. I added the following line to .vimrc to disable colorization: :filetype off Of course, backup config files before you modify them. From gate at ilive4code.net Thu Mar 11 19:38:11 2004 From: gate at ilive4code.net (Greg Sidelinger) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <20040311191925.3e070bf6.jep200305@columbus.rr.com> References: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> <002101c407c2$e01e2ba0$6400a8c0@sundaram> <20040311191925.3e070bf6.jep200305@columbus.rr.com> Message-ID: <1079051891.4600.1.camel@sanitarium> try changing your term type to ansie or ansie-mono. The setting is under Connection and called Terminal-type string. greg On Thu, 2004-03-11 at 19:19, Jim P wrote: > Sundaram Ramasamy wrote: > > > I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do > > ls or commenting the shell, it displays in blue color, it was visible. > > > > How will I change the color? disable the color. > > This will likely take a variety of changes in a variety of places. > > Maybe you can tell putty to tell the remote computer, > that your (putty) terminal is not capable of displaying color. > You would find an old or contrive a new colorless terminal > in termcap on the remote computer and then tell putty to tell > the remote computer that it is that colorless terminal. > Hacking termcap entries (or its replacement) can be intimidating, > but just removing colors should not be too hard. > That would be a rather global change. > > Other changes would be to individually configure each program, > such as ls, perhaps with config files, or alias tricks such as > > alias ls='ls --color=never' > > in ~/.bashrc, /etc/bashrc or /etc/skel/.bashrc, > depending on for how many users you want to make this change for. > ~/.bashrc is probably good enough for you. > > You'll probably want to read the dox for vim > and disable colors or the syntax highlighting that > provokes them in ~/.vimrc. I added the following line to .vimrc > to disable colorization: > > :filetype off > > Of course, backup config files before you modify them. > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > From whooper at freeshell.org Thu Mar 11 20:05:19 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <002101c407c2$e01e2ba0$6400a8c0@sundaram> References: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> <002101c407c2$e01e2ba0$6400a8c0@sundaram> Message-ID: <64660.65.41.50.216.1079053519.squirrel@65.41.50.216> Sundaram Ramasamy said: > Hi, > > I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do > ls or commenting the shell, it displays in blue color, it was visible. > > How will I change the color? disable the color. PuTTY's colors are under "Window -> colours". I personally prefer a white background and leave the rest alone, but if you want you can adjust ANSI Blue to be any color. -- William Hooper From colug at jmglov.net Thu Mar 11 21:35:18 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <002101c407c2$e01e2ba0$6400a8c0@sundaram> References: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> <002101c407c2$e01e2ba0$6400a8c0@sundaram> Message-ID: <20040312023518.GA19769%jmglov@jmglov.net> Quoth Sundaram Ramasamy (Thu 2004-03-11 06:44:50PM -0500): > I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do > ls or commenting the shell, it displays in blue color, it was visible. > > How will I change the color? disable the color. unalias -a Red Hat sets up some aliases, one of which makes ls output in coloured mode. This annoys me, as does their aliasing of rm to 'rm -i'. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040311/74c18738/attachment.bin From rfunk at funknet.net Fri Mar 12 00:46:33 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] changing putty display color In-Reply-To: <20040312023518.GA19769%jmglov@jmglov.net> References: <45225B14-7228-11D8-BEC9-00039317745E@functionalmedia.com> <002101c407c2$e01e2ba0$6400a8c0@sundaram> <20040312023518.GA19769%jmglov@jmglov.net> Message-ID: <200403120046.33368.rfunk@funknet.net> Josh Glover wrote: > Quoth Sundaram Ramasamy (Thu 2004-03-11 06:44:50PM -0500): > > I am using putty to ssh to redhat Linux (8/9/ES and fedora). When I do > > ls or commenting the shell, it displays in blue color, it was visible. > > > > How will I change the color? disable the color. > > unalias -a Or just unalias ls. Or type "\ls" instead of "ls". > Red Hat sets up some aliases, one of which makes ls output in coloured > mode. I only mind color ls when the colors don't work with my terminal colors, and then it's easy to do any number of things to disable it (setting a non-color terminal type probbaly being the best). > This annoys me, as does their aliasing of rm to 'rm -i'. Yeah, I've always hated rm -i too. And it's always set as the user default. I think the idea is that if you know enough to get rid of that alias, you know enough to be careful with rm. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From beano21 at direcway.com Fri Mar 12 02:27:34 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] SAX issue In-Reply-To: <200403120046.33368.rfunk@funknet.net> Message-ID: I have ipgraded my Video card to an Nvidia. Sarted up SuSE and it configured it correctly until it got to SAX. Is there a way to get by this, and have it configured corectly. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From blata at extent0006.entomology.ohio-state.edu Fri Mar 12 08:21:15 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] SAX issue In-Reply-To: References: Message-ID: <4051B94B.4020407@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 are you using the suse drivers or the Nvidia drivers? You wrote this to me on 03/12/2004 02:27 AM: |I have ipgraded my Video card to an Nvidia. Sarted up SuSE and it |configured it correctly until it got to SAX. Is there a way to get by this, |and have it configured corectly. | |--- |Outgoing mail is certified Virus Free. |Checked by AVG anti-virus system (http://www.grisoft.com). |Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 | |_______________________________________________ |colug mailing list |colug@colug.net |http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAUblKv+6+qSFv3TURAiQ9AJ4+aqpjf1CvUQ1n+vyPHBMcOSaAGACfYQwT IvBqnOyhapja9IFHQSEMRO0= =k0XJ -----END PGP SIGNATURE----- From gate at ilive4code.net Fri Mar 12 09:00:22 2004 From: gate at ilive4code.net (Greg Sidelinger) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] SAX issue In-Reply-To: <4051B94B.4020407@bugs.osu.edu> References: <4051B94B.4020407@bugs.osu.edu> Message-ID: <1079100021.4341.40.camel@sanitarium> If you go to nvidias website there is a link with instructions for SUSE. You have to so something like sax -m0: nvidia. But the link seems to be down right now for some reason so I can't check and see exactly what it is. Greg On Fri, 2004-03-12 at 08:21, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > are you using the suse drivers or the Nvidia drivers? > > > > You wrote this to me on 03/12/2004 02:27 AM: > > |I have ipgraded my Video card to an Nvidia. Sarted up SuSE and it > |configured it correctly until it got to SAX. Is there a way to get > by this, > |and have it configured corectly. > | > |--- > |Outgoing mail is certified Virus Free. > |Checked by AVG anti-virus system (http://www.grisoft.com). > |Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > | > |_______________________________________________ > |colug mailing list > |colug@colug.net > |http://www.colug.net/mailman/listinfo/colug > > > - -- > Wade Pinkston > The Ohio State University > Extension Entomology > 1991 Kenny Rd > Columbus OH 43210 > phone: (614) 292-5274 > > Ipsa scientia potestas est > > Windows,a > 32 bit graphical interface for a > 16 bit patch to an > 8 bit operating system internally coded for a > 4 bit processor written by a > 2 bit company that can't stand > 1 bit of competition > > GnuPG Key ID 0x216FDD35 > gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.1 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFAUblKv+6+qSFv3TURAiQ9AJ4+aqpjf1CvUQ1n+vyPHBMcOSaAGACfYQwT > IvBqnOyhapja9IFHQSEMRO0= > =k0XJ > -----END PGP SIGNATURE----- > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > From beano21 at direcway.com Fri Mar 12 09:51:49 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:43 2005 Subject: [COLUG] SAX issue In-Reply-To: <1079100021.4341.40.camel@sanitarium> Message-ID: I am using the nv driver right now. I was going to download the NVidia drivers later. I guess I have to now. -----Original Message----- From: colug-bounces@colug.net [mailto:colug-bounces@colug.net]On Behalf Of Greg Sidelinger Sent: Friday, March 12, 2004 9:00 AM To: Central OH Linux User Group Subject: Re: [COLUG] SAX issue If you go to nvidias website there is a link with instructions for SUSE. You have to so something like sax -m0: nvidia. But the link seems to be down right now for some reason so I can't check and see exactly what it is. Greg On Fri, 2004-03-12 at 08:21, Wade Pinkston wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > are you using the suse drivers or the Nvidia drivers? > > > > You wrote this to me on 03/12/2004 02:27 AM: > > |I have ipgraded my Video card to an Nvidia. Sarted up SuSE and it > |configured it correctly until it got to SAX. Is there a way to get > by this, > |and have it configured corectly. > | > |--- > |Outgoing mail is certified Virus Free. > |Checked by AVG anti-virus system (http://www.grisoft.com). > |Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 > | > |_______________________________________________ > |colug mailing list > |colug@colug.net > |http://www.colug.net/mailman/listinfo/colug > > > - -- > Wade Pinkston > The Ohio State University > Extension Entomology > 1991 Kenny Rd > Columbus OH 43210 > phone: (614) 292-5274 > > Ipsa scientia potestas est > > Windows,a > 32 bit graphical interface for a > 16 bit patch to an > 8 bit operating system internally coded for a > 4 bit processor written by a > 2 bit company that can't stand > 1 bit of competition > > GnuPG Key ID 0x216FDD35 > gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.1 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFAUblKv+6+qSFv3TURAiQ9AJ4+aqpjf1CvUQ1n+vyPHBMcOSaAGACfYQwT > IvBqnOyhapja9IFHQSEMRO0= > =k0XJ > -----END PGP SIGNATURE----- > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.605 / Virus Database: 385 - Release Date: 3/1/2004 From linux at litenverden.org Fri Mar 12 16:59:55 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Reinstall every 7 days? In-Reply-To: <20040127142442.52C8B304021@dalamar.ilive4code.net> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> Message-ID: <405232DB.6020302@litenverden.org> When attempting to find a file (Mandrake 9.2, Gnome 2.4) I get a message that says: "Error while searching for files While searching the following errors were reported: warning:locate:warning:database /varlib/slocate/slocate.db' is more than 8 days old" Of course the error message provides no helpful hint as to what to do next. I could reinstall Linux every 7 days, but this is probably not the preferred course of action. Any thoughts? George > From jim at rossberry.com Fri Mar 12 17:10:21 2004 From: jim at rossberry.com (Jim Wildman) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Reinstall every 7 days? In-Reply-To: <405232DB.6020302@litenverden.org> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> Message-ID: On Fri, 12 Mar 2004, George H. Yeager wrote: > "Error while searching for files > > While searching the following errors were reported: > > warning:locate:warning:database /varlib/slocate/slocate.db' > is more than 8 days old" > > Of course the error message provides no helpful hint as to what to do next. man -k slocate slocate (1) - Security Enhanced version of the GNU Locate slocate [locate] (1) - Security Enhanced version of the GNU Locate updatedb (1) - update the slocate database then man updatedb then ..... ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim@rossberry.com http://www.rossberry.com From peter.king at utoronto.ca Fri Mar 12 17:16:41 2004 From: peter.king at utoronto.ca (Peter King) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Reinstall every 7 days? In-Reply-To: <405232DB.6020302@litenverden.org> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> Message-ID: <20040312221641.GA24067@socrates> On Fri, Mar 12, 2004 at 04:59:55PM -0500, George H. Yeager wrote: > When attempting to find a file (Mandrake 9.2, Gnome 2.4) I get a message > that says: > > "Error while searching for files > > While searching the following errors were reported: > > warning:locate:warning:database /varlib/slocate/slocate.db' > is more than 8 days old" > > Of course the error message provides no helpful hint as to what to do next. > > I could reinstall Linux every 7 days, but this is probably not the > preferred course of action. > > Any thoughts? As root: # updatedb If you get tired of doing this manually -- and who wouldn't -- put it in a cron script to execute regularly. Or you could install once a week. There have been times when it felt like that would have been *less* work than I was already doing... -- Peter King peter.king@utoronto.ca Department of Philosophy 215 Huron Street The University of Toronto (416)-978-3788 ofc Toronto, ON M5S 1A1 CANADA http://individual.utoronto.ca/pking/ ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://www.colug.net/pipermail/colug/attachments/20040312/dc89ab52/attachment.bin From bob at disclosed.org Fri Mar 12 17:22:13 2004 From: bob at disclosed.org (Robert Jewell) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Reinstall every 7 days? In-Reply-To: <405232DB.6020302@litenverden.org> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> Message-ID: <40523815.2080102@disclosed.org> George H. Yeager wrote: > "Error while searching for files > > While searching the following errors were reported: > > warning:locate:warning:database /varlib/slocate/slocate.db' > is more than 8 days old" > > Of course the error message provides no helpful hint as to what to do next. > > I could reinstall Linux every 7 days, but this is probably not the > preferred course of action. > > Any thoughts? > > George heh, no, i think there may be a better solution. :) updatedb is the program that will scan your files and updated that "slocate.db" (which is a file database) .. the whole point of locate is that it's superfast because it's not actually accessing every file on every directory on your harddrive.. it's done that already once and stored it, because file locations don't really change that much. you can run updatedb as root, or ideally set up a cronjob to run it at, say, 4am every night. (it takes a while, depending on the number of files.. at least a few minutes.) (man crontab if you need help on that, or just reply) -bob From linux at litenverden.org Fri Mar 12 17:29:58 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Reinstall every 7 days? In-Reply-To: References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> Message-ID: <405239E6.7030402@litenverden.org> Yep, that did it. I'm not sure I would ever have thought to look for info in the slocate man. I'll have to keep the thinking cap on. Thanks, Jim. George Jim Wildman wrote: > On Fri, 12 Mar 2004, George H. Yeager wrote: > > >>"Error while searching for files >> >>While searching the following errors were reported: >> >>warning:locate:warning:database /varlib/slocate/slocate.db' >>is more than 8 days old" >> >>Of course the error message provides no helpful hint as to what to do next. > > > man -k slocate > slocate (1) - Security Enhanced version of the GNU Locate > slocate [locate] (1) - Security Enhanced version of the GNU Locate > updatedb (1) - update the slocate database > > then man updatedb > > then ..... > > ------------------------------------------------------------------------ > Jim Wildman, CISSP, RHCE jim@rossberry.com > http://www.rossberry.com > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > > From jep200305 at columbus.rr.com Sat Mar 13 22:12:49 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] convertfs Message-ID: <20040313221249.0b6d0df9.jep200305@columbus.rr.com> _do_ backup before trying this. It seems that someone asked about converting from one filesystem to another. >From April LJ, p12: Need to convert from ext2 to something like ReiserFS or XFS? With convertfs, you can do it on the fly on any unmounted filesystem or on one you can unmount from a running system. ... http://tzukanov.narod.ru/convertfs/index.html Which starts off saying: This simple toolset allows you to change type of file system in the lack of backup space. But I will add, you'd better backup your data first if you care about it. From skippy at skippy.net Sun Mar 14 10:54:18 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 Message-ID: <4054802A.2040201@skippy.net> I'm currently running Apache 1.3.28 with PHP 4.3.3 and mod_ssl. I've found a need for mod_rewrite, and may as well upgrade Apache while I'm at it. The question is: do I stick with Apache 1.3.29 or plunge into Apache2 2.0.48? The PHP documentation clearly states not to use PHP and Apache2 in a production environment: http://www.php.net/manual/en/install.apache2.php "Warning Do not use Apache 2.0 and PHP in a production environment neither on Unix nor on Windows." Okay. Obviously lots of people _are_ using this combination on production servers. Should I? What does Apache2 get me? First, mod_ssl is deprecated, with SSL extensions included into the core Apache2 distribution. Apache2 supports "Multi-Processing Modules" (MPMs): http://httpd.apache.org/docs-2.0/mpm.html PHP on Apache2 _requires_ the "prefork" MPM, according to what I read, so I guess this isn't much of a selling point after all. Apache2 has a streamlined configuration file. Cool. What other benefits will I see from stepping up to Apache2? What hidden dangers exist? Should I heed the PHP documentation's advice, and stick with Apache 1.3.29? Thanks, Scott From george at byrd.net Sun Mar 14 11:13:42 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] PHP Database IDE RAD? In-Reply-To: <4054802A.2040201@skippy.net> References: <4054802A.2040201@skippy.net> Message-ID: <405484B6.4000108@byrd.net> Hi All, I am developing web pages with php/apache/mysql and am so tired of having to do everything by scratch. Given that this is one of the most popular combinations for developing webpages, I would have thought that RAD tools would have been developed by now, but I can't find anything after extensive googling and searching of sourceforge/freshmeat that even comes close. Does anyone have any idea if there are any RAD tools, even in development stage, for doing php/mysql work? Ideally, anything even remotely close to dreamweaver or microsoft devstudio where you can drag and drop database fields onto a form, validate forms, etc... Any help or suggestions at all? Thanks, George :) From nordlus at ilive4code.net Sun Mar 14 14:48:24 2004 From: nordlus at ilive4code.net (Stephen Nordlund) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] PHP Database IDE RAD? In-Reply-To: <405484B6.4000108@byrd.net> References: <4054802A.2040201@skippy.net> <405484B6.4000108@byrd.net> Message-ID: <1070.24.92.129.240.1079293704.squirrel@mail.ilive4code.net> I have wondered the same thing. Here is what I have come up with: http://nordlund.ilive4code.net/phpcodegenie1.4/ phpCodeGenie is a code writer for PHP/MySQL applications. Just design your database tables and phpCodeGenie can write the php scripts and programs for you. phpCodeGenie will build data entry forms, insert scripts, database lister scripts, edit record forms, update record scripts, delete confirmation scripts, delete scripts, search forms, search scripts and other frontend - database interaction code. Basically, it provides a core basic code from which you can build on. I hope that you enjoy ! http://nordlund.ilive4code.net/phpform/ phpFormGenerator is an easy-to-use tool to create reliable and efficient web forms. No programming of any sort is required. It has the ability to create up to 100 form fields, and to add a variety of field types including text boxes, drop down selection m **Note this next app is not fully functional. I justed put the link on here as a bit of a teaser I suppose.. I'm just starting to get into the MVC aspect of OOP. http://nordlund.ilive4code.net/phpmvc/ php.MVC implements the Model-View-Controller (MVC) design pattern, and encourages application design based on the Model 2 paradigm. This design model allows the Web page or other contents (View) to be mostly separated from the internal application code (Controller/Model), making it easier for designers and programmers to focus on their respective areas of expertise. The framework provides a single entry point Controller. The Controller is responsible for allocating HTTP requests to the appropriate Action handler (Model) based on configuration mappings. The Model contains the business logic for the application. The Controller then forwards the request to the appropriate View component, which is usually implemented using a combination of HTML with PHP tags in the form of templates. The resulting contents are returned to the client browser, or via another protocol such as SMTP. php.MVC is a PHP port of Jakarta Struts. It currently supports many features of Struts, including declarative application configuration via the XML digester. For example, mappings from the various Action business logic components to appropriate results pages can be specified declaratively in the XML configuration file. Hope this helps.. -- Stephen Nordlund, MCP Systems Engineer George Byrd said: > Hi All, > > I am developing web pages with php/apache/mysql and am so tired of > having to do everything by scratch. Given that this is one of the most > popular combinations for developing webpages, I would have thought that > RAD tools would have been developed by now, but I can't find anything > after extensive googling and searching of sourceforge/freshmeat that > even comes close. > > Does anyone have any idea if there are any RAD tools, even in > development stage, for doing php/mysql work? > > Ideally, anything even remotely close to dreamweaver or microsoft > devstudio where you can drag and drop database fields onto a form, > validate forms, etc... > > Any help or suggestions at all? > > Thanks, > > George :) > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > From rutledge.50 at osu.edu Mon Mar 15 11:14:53 2004 From: rutledge.50 at osu.edu (LINCOLN RUTLEDGE) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Fwd: [FWD: software vendor agreement] Message-ID: Forwarded to the list for a friend (netscape mail blocked by the list server?): Has anyone on the list had to sign an agreement at work to only use Microsoft software? I was asked to sign one that said that I was free to make suggestions as to any changes to make to any servers or network configuration. However it explicitly stated that I am not to make any changes anywhere with any equipment that conflicts with said party's choice of Microsoft as a vendor. I have never heard of anyone having to do something like this, I am shocked. I like to think of myself as mature and I haven't started a flame on a newsgroup or mailing list for years. But I cannot believe this. I have been studying Linux and the BSDs for years. I am self taught and make a good living in the business world wrangling computers, servers, and networks. Most of the knowledge and abilities I have now is because of the "hands-on imperative" techniques inspired by Linux when it comes to systems and networking. The open source code and documentation provides a transparent window into the machine that has allowed me to get where I am and do what I do today. I don't want to say too much because I don't want to lose my job or screw up my career, which is why I'm using this silly email address. I am very indifferent to what I have to work with to make a decent living, but the SCO memo and now this, I'm starting to think Big Brother has it out for me. confused, a8u5ed8ym5 From pstjohn at hrblock.com Mon Mar 15 12:13:13 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Fwd: [FWD: software vendor agreement] Message-ID: SCO started out as a subsidiary of Microsoft, got spun off, then bought into the "unix" licensing with investment capital from Microsoft. Your problem isn't with Big Brother, it's with Microsoft; which is monopolistic. This is my opinion and does not reflect the position of my employer. Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of LINCOLN RUTLEDGE > Sent: Monday, March 15, 2004 11:15 AM > To: colug@colug.net > Subject: [COLUG] Fwd: [FWD: software vendor agreement] > > Forwarded to the list for a friend (netscape mail blocked by the list server?): > > Has anyone on the list had to sign an agreement at work to only use Microsoft software? I was asked to sign one that said that I was free to make suggestions as to any changes to make to any servers or network configuration. However it explicitly stated that I am not to make any changes anywhere with any equipment that conflicts with said party's choice of Microsoft as a vendor. > > I have never heard of anyone having to do something like this, I am shocked. I like to think of myself as mature and I haven't started a flame on a newsgroup or mailing list for years. But I cannot believe this. I have been studying Linux and the BSDs for years. I am self taught and make a good living in the business world wrangling computers, servers, and networks. Most of the knowledge and abilities I have now is because of the "hands-on imperative" techniques inspired by Linux when it comes to systems and networking. The open source code and documentation provides a transparent window into the machine that has allowed me to get where I am and do what I do today. > > I don't want to say too much because I don't want to lose my job or screw up my career, which is why I'm using this silly email address. > > I am very indifferent to what I have to work with to make a decent living, but the SCO memo and now this, I'm starting to think Big Brother has it out for me. > > confused, > > a8u5ed8ym5 > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040315/a96081b8/attachment.htm From WKehr at checkfree.com Mon Mar 15 13:18:20 2004 From: WKehr at checkfree.com (WKehr@checkfree.com) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Fwd: [FWD: software vendor agreement] In-Reply-To: Message-ID: Does this mean that any plugin to IE which is not written by Microsoft is disallowed? If a printer is purchased, that the manufacturer's drivers cannot be used? How does one view a PDF? How can you boot your computer with a non-Microsoft bios? You must disable Java-script or active-x because these are basically programs written by non-Microsoft people. In fact, you cannot run any of the programs written at the place where you work. LINCOLN RUTLEDGE Sent by: colug-bounces@colug.net 03/15/2004 11:14 AM Please respond to Central OH Linux User Group To colug@colug.net cc Subject [COLUG] Fwd: [FWD: software vendor agreement] Forwarded to the list for a friend (netscape mail blocked by the list server?): Has anyone on the list had to sign an agreement at work to only use Microsoft software? I was asked to sign one that said that I was free to make suggestions as to any changes to make to any servers or network configuration. However it explicitly stated that I am not to make any changes anywhere with any equipment that conflicts with said party's choice of Microsoft as a vendor. I have never heard of anyone having to do something like this, I am shocked. I like to think of myself as mature and I haven't started a flame on a newsgroup or mailing list for years. But I cannot believe this. I have been studying Linux and the BSDs for years. I am self taught and make a good living in the business world wrangling computers, servers, and networks. Most of the knowledge and abilities I have now is because of the "hands-on imperative" techniques inspired by Linux when it comes to systems and networking. The open source code and documentation provides a transparent window into the machine that has allowed me to get where I am and do what I do today. I don't want to say too much because I don't want to lose my job or screw up my career, which is why I'm using this silly email address. I am very indifferent to what I have to work with to make a decent living, but the SCO memo and now this, I'm starting to think Big Brother has it out for me. confused, a8u5ed8ym5 _______________________________________________ colug mailing list colug@colug.net http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040315/22f09748/attachment.htm From jonadab at bright.net Tue Mar 16 10:55:16 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Fwd: [FWD: software vendor agreement] In-Reply-To: References: Message-ID: <4qso7pnf.fsf@jonadab.homeip.net> LINCOLN RUTLEDGE writes: > I was asked to sign one that said that I was free to make > suggestions as to any changes to make to any servers or network > configuration. However it explicitly stated that I am not to make > any changes anywhere with any equipment that conflicts with said > party's choice of Microsoft as a vendor. My advice would be, only sign it if you intend to follow it. I don't mean that this has to have been your original intention, but if you decide to sign the thing, you need to abide by it or get another job. If you can quietly avoid signing it (by, say, letting it get burried under a pile of other papers and never getting around to it), that *might*, depending on your work environment, be better than raising the issue and explaining why you aren't willing to sign it. (Then again, it might not; depends on your work environment. Also might depend on whose idea the thing was originally.) -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From 6f at earthlink.net Tue Mar 16 12:53:24 2004 From: 6f at earthlink.net (Steve Ferguson) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Sound In-Reply-To: <200403060056.07823.rfunk@funknet.net> References: <20040306045723.GB5485@earthlink.net> <200403060056.07823.rfunk@funknet.net> Message-ID: <20040316175324.GA7490@earthlink.net> Finally got it working. Thanks to all! Steve On Sat, Mar 06, 2004 at 12:56:07AM -0500, Rob Funk wrote: > Steve Ferguson wrote: > > I can't get any audio to work with XMMS (or anything else for that > > matter). I've tried all combinations of > > plugins, etc. Any ideas? Anyone work with this driver before? > > Did you check the mixer and make sure the volume is up and not muted? > > I can't tell whether you're using OSS or ALSA, but I believe ALSA defaults > to muted. > > -- > ==============================| "A slice of life isn't the whole cake > Rob Funk | One tooth will never make a full grin" > http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug From george at byrd.net Tue Mar 16 13:30:52 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:44 2005 Subject: [COLUG] Sound In-Reply-To: <20040316175324.GA7490@earthlink.net> References: <20040306045723.GB5485@earthlink.net> <200403060056.07823.rfunk@funknet.net> <20040316175324.GA7490@earthlink.net> Message-ID: <405747DC.2080907@byrd.net> Hi Steve, Any advice on what the problem was or how you fixed it? LOTS of people have problems with sound, so any information to go into the archives I'm sure would help people. Thanks, and congratulations! George :) Steve Ferguson wrote: >Finally got it working. Thanks to all! > >Steve > >On Sat, Mar 06, 2004 at 12:56:07AM -0500, Rob Funk wrote: > > >>Steve Ferguson wrote: >> >> >>>I can't get any audio to work with XMMS (or anything else for that >>>matter). I've tried all combinations of >>>plugins, etc. Any ideas? Anyone work with this driver before? >>> >>> >>Did you check the mixer and make sure the volume is up and not muted? >> >>I can't tell whether you're using OSS or ALSA, but I believe ALSA defaults >>to muted. >> >>-- >>==============================| "A slice of life isn't the whole cake >> Rob Funk | One tooth will never make a full grin" >> http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" >> >>_______________________________________________ >>colug mailing list >>colug@colug.net >>http://www.colug.net/mailman/listinfo/colug >> >> >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > > > > From rfunk at funknet.net Tue Mar 16 13:46:26 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <4054802A.2040201@skippy.net> References: <4054802A.2040201@skippy.net> Message-ID: <200403161346.26912.rfunk@funknet.net> Scott Merrill wrote: > The question is: do I stick with Apache 1.3.29 or plunge into > Apache2 2.0.48? Stick with 1.3. 1.3 is highly stable and much better tested. Apache 2 isn't ready for prime time, at least if you go beyond what's included with it. > The PHP documentation clearly states not to use PHP and Apache2 in a > production environment: > http://www.php.net/manual/en/install.apache2.php > "Warning > > Do not use Apache 2.0 and PHP in a production environment neither on > Unix nor on Windows." > > Okay. Obviously lots of people _are_ using this combination on > production servers. Should I? The PHP people say don't do it. If you want a stable server, follow their advice. > What does Apache2 get me? That's the real question. The main change in Apache 2 is that it does threading rather than forking. This is an advantage on Windows, but not on Unix. On Unix it's just a huge disruption that causes more problems than it solves. ESR has a good discussion of threading as it relates to Unix in The Art Of Unix Programming. http://www.catb.org/~esr/writings/taoup/html/ch07s03.html#id2923889 -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From tom at functionalmedia.com Tue Mar 16 14:51:45 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <200403161346.26912.rfunk@funknet.net> Message-ID: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> >> What does Apache2 get me? > > That's the real question. The main change in Apache 2 is that it does > threading rather than forking. This is an advantage on Windows, but > not > on Unix. On Unix it's just a huge disruption that causes more problems > than it solves. > ESR has a good discussion of threading as it relates to Unix in The > Art Of > Unix Programming. > http://www.catb.org/~esr/writings/taoup/html/ch07s03.html#id2923889 > Just some thoughts. My understanding with apache 2 is that it allows threading if the underlying OS supports it. You can still configure with-mpm=prefork to get the 1.3 style preforking functionality. I read the Raymond article on threads as a threat and I agree but one of the goals in the push for apache 2.0 was better support for non-unix operating systems. Perhaps that was the mistake, if indeed there was a mistake. Making big modules like PHP threadsafe has been time consuming. Please understand that I am not a programmer just a user. Other features of apache 2.0 would be the use of separate userid's for different virtual hosts. The promise of output filters, although I suppose that sending a request through two filters is asking for trouble. Running a CGI that also has SSI(server side includes) for example. It is nice to have mod_ssl as a standard module, As I recall in 1.3 it was separate. So in exchange for threads we get better functionality on windows, and a few other features. The negative is that all modules need to be thread safe. As a non programmer I do not have a grasp on how big of a headache this is. Seems like it was stalled in 1.3 for a long time. IBM was behind the push for 2.0. Apache is bundled in there websphere(?) product. Tom -- Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From jim at rossberry.com Tue Mar 16 15:07:06 2004 From: jim at rossberry.com (Jim Wildman) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> Message-ID: On Tue, 16 Mar 2004, tom hanlon wrote: > Seems like it was stalled in 1.3 for a long time. IBM was behind the > push for 2.0. Apache is bundled in there websphere(?) product. My impression is that the 2.0 model makes porting from IIS much easier as well. I've seen several commercial products that perform well in their IIS and Apache 2.0 incarnations and poorly on Apache 1.x. That would explain IBM. ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim@rossberry.com http://www.rossberry.com From rfunk at funknet.net Tue Mar 16 15:11:19 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> Message-ID: <200403161511.19180.rfunk@funknet.net> tom hanlon wrote: > one of the goals in the push for > apache 2.0 was better support for non-unix operating systems. Sure, but that has little impact on your decision if you're running Unix. > Perhaps > that was the mistake, if indeed there was a mistake. Making big modules > like PHP threadsafe has been time consuming. Also mod_perl has had problems stabilizing for Apache 2. > It is nice to have mod_ssl as a standard module, As I recall > in 1.3 it was separate. Simple matter of packaging. Various distributions make it easy to get an SSL-enable Apache. (It's a standard part of OpenBSD too.) > So in exchange for threads we get better functionality on windows, ... which doesn't matter to Unix sysadmins.... > and > a few other features. Most of which nobody will use anytime soon. (Not counting packaging issues as features.) > The negative is that all modules need to be > thread safe. As a non programmer I do not have a grasp on how big of a > headache this is. Huge headache. > Seems like it was stalled in 1.3 for a long time. What one person sees as "stalled", another sees as "stability". Particularly when bug fixes have been released regularly for 1.3. A third person would say "it's just an arbitrary number". > IBM was behind the > push for 2.0. Apache is bundled in there websphere(?) product. Which is great if you get IBM's support. There's a good reason for the computer industry's aversion to "dot-zero" releases. And Apache 2 is still a dot-zero release. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From s.molnar at sbcglobal.net Tue Mar 16 16:31:20 2004 From: s.molnar at sbcglobal.net (Stephen P. Molnar, Ph.D.) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Large File on Boot Drive - What is It? Message-ID: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> I have a rather large file, 402,653,184 byters to be exact, Pagefile.sys. It has the attributes hsa. What might it be and is it safe to remove it? Thanks in advance From paul.bender at acm.org Tue Mar 16 16:33:52 2004 From: paul.bender at acm.org (paul.bender@acm.org) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Large File on Boot Drive - What is It? In-Reply-To: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> Message-ID: <200403162134.i2GLYUtP001740@ms-smtp-02-eri0.ohiordc.rr.com> On 16 Mar, Stephen P. Molnar, Ph.D. wrote: > I have a rather large file, 402,653,184 byters to be exact, > Pagefile.sys. It has the attributes hsa. What might it be and is it > safe to remove it? On a Windows based computer, Pagefile.sys is the file Windows uses for virtual memory. Windows might recreate it if it disappears, so It might be safe to delete it, but it might come back as well. If this is on a Linux only box, It's probably safe to delete it. Paul -- ______________________________________________________________________________ "Quality is a Characteristic of thought and statement that is recognized by a nonthinking process. Because definitions are a product of rigid formal thinking, quality cannot be defined." Robert M. Pirsig Zen and The Art of Motorcycle Maintenance -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040316/3c8fc342/attachment.bin From jep200305 at columbus.rr.com Tue Mar 16 16:39:45 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Large File on Boot Drive - What is It? In-Reply-To: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> References: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> Message-ID: <20040316163945.5a63097d.jep200305@columbus.rr.com> "Stephen P. Molnar, Ph.D." wrote: > I have a rather large file, 402,653,184 byters to be exact, > Pagefile.sys. It has the attributes hsa. What might it be and is it safe > to remove it? Is this on a Windows box? From pstjohn at hrblock.com Tue Mar 16 16:39:55 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Large File on Boot Drive - What is It? Message-ID: Just to get you a quick answer: no, I don't think it's safe, I think that might be virutal memory. You might check who owns it, the permisions; but I'm sure a flavor-specif linux guru will give you a better answer shortly. Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Stephen P. Molnar, Ph.D. > Sent: Tuesday, March 16, 2004 4:31 PM > To: colug@colug.net > Subject: [COLUG] Large File on Boot Drive - What is It? > Importance: High > > I have a rather large file, 402,653,184 byters to be exact, > Pagefile.sys. It has the attributes hsa. What might it be and is it safe > to remove it? > > Thanks in advance > > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040316/5bd2bf5c/attachment.htm From colug at jmglov.net Tue Mar 16 17:07:32 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <200403161511.19180.rfunk@funknet.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403161511.19180.rfunk@funknet.net> Message-ID: <3834.24.123.50.150.1079474852.squirrel@mail.jmglov.net> Quoth Rob Funk: > tom hanlon wrote: > >> The negative is that all modules need to be >> thread safe. As a non programmer I do not have a grasp on how big of a >> headache this is. > > Huge headache. In my experience, code that is hard to make thread-safe is not so well designed. For those who are not familiar with the difference between a multi-threaded program and a multi-process one, it boils down to this: Threads share the same address space, and child processes do not. [1] What this means it that multi-process apps must communicate via IPC (Inter Process Communication) methods such as pipes, shared memory, the filesystem, etc, whereas multi-threaded apps can communicate via data structures. However, this ability to share data between threads means race conditions emerge. A race condition is a point in your program where two (or more) separate threads are attempting to access the same data structure. Consider this simple example: [2] state = machines[i]->state; state++; machines[i]->state = state; Suppose thread A and thread B both attempt to execute this line of code. Say machines[i]->state starts out as 3. One possible outcome is that thread A fetches the state (state == 3), adds one to it (state == 4), and stores it (state == 4). Then thread B does the same thing, so machines[i]->state ends up as 5. But in multi-threaded programming, you must consider *all possible interleavings* of non-atomic instructions. So another possible outcome is this: A: state = machines[i]->state; // 3 B: state = machines[i]->state; // 3 A: state++; // 4 A: machines[i]->state = state; // 4 B: state++; // 4 B: machines[i]->state = state; // 4 This is a very simple example of a race, but one that illustrates the basic concept. To prevent races, you must have some way of locking a variable (or data structure) so that one thread can execute a series of instructions that must be atomic. Luckily, modern operating systems give us synchronisation constructs that allow us to write thread-safe code. Writing thread-safe code is a huge pain when you have a lot of global data structures. If you write modular code, you will find thread safety not so difficult to implement. In fact, Java makes it down right simple: there is a keyword ("synchronized", IIRC) that you add to a method and then the mutex is done for you. Multi-threaded code *is* faster, if done right, because it tends to keep the data structures "cache-hot", and IPC has more overhead than direct communication. -Josh [1] Of course, Linux (and probably other Unices) has a performance hack called "copy on write", where child processes start out sharing the parent's address space until either the parent or the child writes to that space, at which time a separate space is created for the child. However, this does not change the way multi-process programs work. [2] Which could also be written as machines[i]->state++, but that does nothing to alleviate the race. Three lines of code just makes the race easier to see. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 From rfunk at funknet.net Tue Mar 16 18:11:51 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <3834.24.123.50.150.1079474852.squirrel@mail.jmglov.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403161511.19180.rfunk@funknet.net> <3834.24.123.50.150.1079474852.squirrel@mail.jmglov.net> Message-ID: <200403161811.51208.rfunk@funknet.net> Josh Glover wrote: > In my experience, code that is hard to make thread-safe is not so well > designed. Depends. Certainly good design contributes to minimizing cases where thread-safety would be a problem. But there are plenty of cases where the thread-safe design is too heavy and complex when thread-safety isn't an issue. The issue I'm looking at, though, is converting a body of stable existing code that was written with certain assumptions in mind, and suddenly needing to change those basic assumptions. That is a huge pain. Certainly it's possible, but it'll take time before the new code can be considered stable. > Multi-threaded code *is* faster, if done right, because it tends to keep > the data structures "cache-hot", and IPC has more overhead than direct > communication. Seems to me that making a major structural change in an attempt to leverage a few hundred kilobytes of cache is a net loss. Don't forget locking shared data, which detracts from the performance. And don't forget the hassle of *remembering* to always properly lock shared data. And if you have no shared data to lock, then you've lost that "cache-hot" advantage. I'm extremely reluctant to trade programmer simplicity (and therefore reliability) for a questionable performance gain until the reliability has been proven. And if you include external modules (mod_perl, mod_php), that hasn't happened yet with Apache 2. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From rfunk at funknet.net Tue Mar 16 18:31:57 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <3834.24.123.50.150.1079474852.squirrel@mail.jmglov.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403161511.19180.rfunk@funknet.net> <3834.24.123.50.150.1079474852.squirrel@mail.jmglov.net> Message-ID: <200403161831.57730.rfunk@funknet.net> Josh Glover wrote: > Multi-threaded code *is* faster, if done right, because it tends to keep > the data structures "cache-hot", and IPC has more overhead than direct > communication. Actually, I'd like to question this some more. There are many types of IPC. While socket-based IPC is certainly slower, shared memory is not, and if i remember correctly Apache 1.3 uses shared memory. With shared memory you get the same "cache-hot" advantages and locking problems as you do with threading, but you don't need to worry about memory that you don't explicitly share. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jonadab at bright.net Tue Mar 16 22:26:28 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> Message-ID: tom hanlon writes: > As a non programmer I do not have a grasp on how big of a headache > this is. I'm just enough of a programmer to have a pretty good idea. Basically, how hard it is to make any given application threadsafe depends very much on the codebase of that application and how it goes about doing certain kinds of things. Where this gets to be a real bummer is if the application relies heavily on a lot of third-party modules that are produced and maintained by people _other_ than the developers of the application itself. You have to do them all. It's a coordination nightmare. Apache gets hit pretty hard by this, because a sizeable portion of the userbase considers Apache to be practically useless without some module or another -- mod_ssl, mod_perl, mod_php, and so on. Some of the modules will be (relatively) easy to make threadsafe, but some of them will not, and this situation is aggravated if the maintainers of some of the modules don't have a solid understanding of the issue (due, presumably, to a lack of experience with threaded programming, which is fairly common still at this point even among otherwise fairly experienced developers). Perl is still going through the conversion to threadsafe programming. The core has been threadsafe for some while now (since 2.5.something I think), but there are still popular modules on the CPAN that are not entirely threadsafe, and so the default build options for Perl still build without threads (and, consequently, the vast overwhelming majority of Perl programmers don't use threads much, if at all). This is largely a result of getting hit by the same problem that is making the transition hard for Apache -- you have to get N random module maintainers who may or may not care at all about threads to all make their code threadsafe. (If anything, this problem hits Perl even harder than it hits Apache, because there are a larger number and variety of popular modules. Heck, there are still modules on the CPAN that won't run under Taint checking, and that's *easy* to fix.) -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From rfunk at funknet.net Tue Mar 16 23:00:12 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> Message-ID: <200403162300.12320.rfunk@funknet.net> Jonadab the Unsightly One wrote: > Apache gets hit pretty hard by this, because a sizeable portion of the > userbase considers Apache to be practically useless without some > module or another -- mod_ssl, mod_perl, mod_php, and so on. Good point. > Perl is still going through the conversion to threadsafe programming. And people using mod_perl get hit with this when they try to convert to Apache 2. Even if both Apache and mod_perl are totally thread-safe (not something I'm willing to bet on yet), there's still the perl modules to deal with. I wouldn't be surprised if PHP is going through something similar. And if the PHP developers say don't use PHP with Apache 2 in production yet, that alone is enough to convince me to hold off. > The core has been threadsafe for some while now (since 2.5.something I > think), Er, there's no such thing as Perl 2.5.something, at least not currently extant and certainly not with threads. :-) I think it's more like 5.6.something. > (and, consequently, the vast overwhelming > majority of Perl programmers don't use threads much, if at all). Well, another reason most perl programmers don't use threads is that they simply don't see the point. Threads are more common in the Windows culture (where forking a new process is expensive and ) than in the Unix culture whence Perl came. More from ESR on threads: http://www.catb.org/~esr/writings/taoup/html/ch07s01.html "Threading is a performance hack.... they do not reduce global complexity but rather increase it, and should therefore be avoided save under dire necessity." and on shared memory: http://www.catb.org/~esr/writings/taoup/html/ch07s02.html#id2923100 "The issues here resemble those in multithreading... but are more manageable because default is not to share memory. Thus, problems are better contained." > This > is largely a result of getting hit by the same problem that is making > the transition hard for Apache -- you have to get N random module > maintainers who may or may not care at all about threads to all make > their code threadsafe. Thanks for the insight. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jonadab at bright.net Tue Mar 16 23:05:56 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Large File on Boot Drive - What is It? In-Reply-To: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> References: <6.0.0.22.2.20040316162846.03502260@pop.sbcglobal.yahoo.com> Message-ID: <65d43yor.fsf@jonadab.homeip.net> "Stephen P. Molnar, Ph.D." writes: > I have a rather large file, 402,653,184 byters to be exact, > Pagefile.sys. It has the attributes hsa. What might it be and is it > safe to remove it? Hmmm, "hsa.", this must be on a FAT filesystem. That plus the .sys extension and I'm pretty sure this is a Windows system. If it's Windows 95, 98, or Me, then it's safe to delete it when the system is not running (e.g., you can boot in command-prompt-only mode and delete it, boot from floppy and delete it, boot another OS and delete it, whatever). But I think 95/98/Me last I checked uses a different filename (win386.swp IIRC), so this is probably NT/2000/XP, in which case I don't know for certain; I think it handles virtual memory a little differently than 95/98/Me but I don't know the details. Search the MS Knowledge Base. If this is not a Windows system, then you need to explain what OS you're using (Linux? What distro?), what filesystem (UMSDOS?) and what directory the file is in. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Wed Mar 17 00:03:56 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <200403162300.12320.rfunk@funknet.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403162300.12320.rfunk@funknet.net> Message-ID: Rob Funk writes: > Er, there's no such thing as Perl 2.5.something, at least not currently > extant and certainly not with threads. :-) I think it's more like > 5.6.something. I meant 5.5.something (the development releases preceding 5.6.0). Not sure where the 2 came from; it's evening and I'm sleepy. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From colug at jmglov.net Wed Mar 17 09:39:55 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:45 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <200403162300.12320.rfunk@funknet.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403162300.12320.rfunk@funknet.net> Message-ID: <4418.24.123.50.150.1079534395.squirrel@mail.jmglov.net> Quoth Rob Funk: > Well, another reason most perl programmers don't use threads is that they > simply don't see the point. Threads are more common in the Windows > culture (where forking a new process is expensive and ) than in the Unix > culture whence Perl came. > > More from ESR on threads: > http://www.catb.org/~esr/writings/taoup/html/ch07s01.html > "Threading is a performance hack.... they do not reduce global complexity > but rather increase it, and should therefore be avoided save under dire > necessity." I don't often disagree with ESR on Unix-related issues, but here I do. I actually consider multi-threaded programming to be *simpler* than multi-process programming. Fork / exec brings with it a whole slew of issues, it is just that Unix programmers of ESR's calibre have been doing the fork / exec dance since the beginning of time (or at least, the beginning of the Unix epoch). With threads, you *don't* have to worry about open filehandles persisting through a fork (which can be a real issue when you have lots of filehandles open). You don't have to go through elaborate IPC hoops. I think that the new generation of programmers, at least those with Computer Science degrees, have just as much knowledge about multi-threaded programming and the issues that surround it as they do about fork / exec-style programming. The bottom line is, keep complexity down. For me, threads are less complex. For you, this might not be the case. We should both code accordingly. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 From mharrold at cas.org Wed Mar 17 09:53:38 2004 From: mharrold at cas.org (Mike Harrold) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <4418.24.123.50.150.1079534395.squirrel@mail.jmglov.net> from "Josh Glover" at Mar 17, 2004 09:39:55 AM Message-ID: <200403171453.i2HErcE25841@mah21awu.cas.org> Josh Glover wrote: > > Quoth Rob Funk: > > > Well, another reason most perl programmers don't use threads is that they > > simply don't see the point. Threads are more common in the Windows > > culture (where forking a new process is expensive and ) than in the Unix > > culture whence Perl came. > > > > More from ESR on threads: > > http://www.catb.org/~esr/writings/taoup/html/ch07s01.html > > "Threading is a performance hack.... they do not reduce global complexity > > but rather increase it, and should therefore be avoided save under dire > > necessity." > > I don't often disagree with ESR on Unix-related issues, but here I do. I > actually consider multi-threaded programming to be *simpler* than > multi-process programming. Fork / exec brings with it a whole slew of issues, > it is just that Unix programmers of ESR's calibre have been doing the fork / > exec dance since the beginning of time (or at least, the beginning of the Unix > epoch). > > With threads, you *don't* have to worry about open filehandles persisting > through a fork (which can be a real issue when you have lots of filehandles > open). You don't have to go through elaborate IPC hoops. > > I think that the new generation of programmers, at least those with Computer > Science degrees, have just as much knowledge about multi-threaded programming > and the issues that surround it as they do about fork / exec-style > programming. > > The bottom line is, keep complexity down. For me, threads are less complex. > For you, this might not be the case. We should both code accordingly. And this is precisely why programming can be such a pain in the ass. Why? Because invariably we end up having to maintain each other's code. We know little to nothing about the way the other person's mind works, there's next to no documentation, and so we end up rewriting huge chunks of code in a way that we are familiar with. So much for code reuse, eh? Personally my biggest problem with multi-threaded code is that invariably it doesn't need to be threaded (i.e., there are no performance gains), and it's simply threaded because the original programmer wanted it to be threaded. IMHO, of course. /Mike From pstjohn at hrblock.com Wed Mar 17 10:51:11 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 Message-ID: In MS Outlook, I can't read the mail from "Jonadab the Unsightly" apparently because "the message flag must be 100 bytes or less". Is there a convenient workaround? Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Rob Funk > Sent: Tuesday, March 16, 2004 11:00 PM > To: Central OH Linux User Group > Subject: Re: [COLUG] Apache 1.3.29 vs Apache 2.0.48 > > Jonadab the Unsightly One wrote: > > Apache gets hit pretty hard by this, because a sizeable portion of the > > userbase considers Apache to be practically useless without some > > module or another -- mod_ssl, mod_perl, mod_php, and so on. ... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040317/ff6a76f3/attachment.htm From jep200305 at columbus.rr.com Wed Mar 17 11:00:44 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: References: Message-ID: <20040317110044.09986d0c.jep200305@columbus.rr.com> "St. Patrick" wrote: > In MS Outlook, I can't read the mail from "Jonadab the Unsightly" > apparently because "the message flag must be 100 bytes or less". > > Is there a convenient workaround? There are several for Windows. Mozilla (mozilla.org) Thunderbird (http://mozilla.org/products/thunderbird/) Sylpheed (http://sylpheed-claws.sourceforge.net/win32/) (My one experience using Sylpheed under Win98SE failed) From pat at linuxcolumbus.com Wed Mar 17 11:05:02 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: References: Message-ID: <20040317160502.GH17638@linuxcolumbus.com> On Wed, Mar 17, 2004 at 09:51:11AM -0600, St. John, Peter wrote: > In MS Outlook, I can't read the mail from "Jonadab the Unsightly" apparently because "the message flag must be 100 bytes or less". > > Is there a convenient workaround? Sure. Reboot and install Linux. :) Pat From rfunk at funknet.net Wed Mar 17 11:28:40 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] threading In-Reply-To: <4418.24.123.50.150.1079534395.squirrel@mail.jmglov.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com> <200403162300.12320.rfunk@funknet.net> <4418.24.123.50.150.1079534395.squirrel@mail.jmglov.net> Message-ID: <200403171128.40626.rfunk@funknet.net> Josh Glover wrote: > Fork / exec brings with it a whole slew of > issues, it is just that Unix programmers of ESR's calibre have been > doing the fork / exec dance since the beginning of time (or at least, > the beginning of the Unix epoch). I don't see how forking has more issues than threading. And exec is a separate issue. I'm not talking about execing another program. > With threads, you *don't* have to worry about open filehandles > persisting through a fork (which can be a real issue when you have lots > of filehandles open). I don't know what you're talking about here. Last I checked, threads share filehandles. And fork doesn't work any differently from a threaded than any other process. (Maybe you're thinking of Windows or some other platform rather than Unix?) > You don't have to go through elaborate IPC hoops. See my previous comments about shared memory. Nothing more elaborate than what you go through with threads is necessary. > I think that the new generation of programmers, at least those with > Computer Science degrees, have just as much knowledge about > multi-threaded programming and the issues that surround it as they do > about fork / exec-style programming. Probably more, in fact, which explains some of your misconceptions. :-) And this is mostly because those of us who took programming classes ten years ago mostly learned on Unix, while those taking programming classes now more often learn on Windows or Java. On those platforms, threading is the way to go since forking a process is more expensive and works differently from how it does in Unix. > The bottom line is, keep complexity down. For me, threads are less > complex. For you, this might not be the case. We should both code > accordingly. My first inclination is to agree with this, but then I remember that at some point I'll have to maintain code written by you or someone like you. :-) I think an essential part of the bottom line is to code in a style appropriate to the platform. In Windows or Java, threads are usually the way to go. But on Unix, threads are usually not the way to go. Oh yeah, and there are times in Unix when forking a process isn't necessarily the right thing, but threading still isn't. That's where an event loop may be useful. This is the method used by the X server, the fastest web servers (e.g. thttpd, Zeus; for details see http://www.acme.com/software/thttpd/benchmarks.html), and other programs that need high-speed handling of many interactions simultaneously. See this presentation by John Ousterhout (inventor of tcl/tk) comparing threads with event loops: http://home.pacbell.net/ouster/threads.pdf I recently worked on a simple example of the event-loop technique: http://www.ex-parrot.com/~pdw/slimp3slave/ -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From rfunk at funknet.net Wed Mar 17 11:30:45 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: References: Message-ID: <200403171130.45240.rfunk@funknet.net> St. John, Peter wrote: > In MS Outlook, I can't read the mail from "Jonadab the Unsightly" > apparently because "the message flag must be 100 bytes or less". Heh. X-message-flag: This message is enhanced with advanced features that this version of Outlook does not support. Please upgrade to the latest version. > Is there a convenient workaround? Junk Outlook. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jeffrey at tadlocks.net Wed Mar 17 11:53:52 2004 From: jeffrey at tadlocks.net (Jeffrey Tadlock) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Message Flag In-Reply-To: References: Message-ID: <20040317165352.GA30822@tadlocks.net> On Wed, Mar 17, 2004 at 09:51:11AM -0600, St. John, Peter wrote: > In MS Outlook, I can't read the mail from "Jonadab the Unsightly" apparently because "the message flag must be 100 bytes or less". > It seems you are getting this because Jonadab has placed an X-message-flag: in his mail headers that is too long for Outlook to process. The message in his header is this: "This message is enhanced with advanced features that this version of Outlook does not support. Please upgrade to the latest version." X-message-flag is processed by the Outlook mail client and is a "fun" way to flag an email message and give Outlook users a "special" message. In the corporate world usually used by users as the Follow-Up flag in Outlook to remind the recipient to follow up on an action or email. Of course it can also be used to suggest to Outlook users to upgrade their mail client to Mozilla, or your client of choice... > Is there a convenient workaround? Not sure there is a work around short of not using Outlook since Outlook seems to only handle 100bytes in the field in question. /jft From colug at jmglov.net Wed Mar 17 12:51:47 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] threading In-Reply-To: <200403171128.40626.rfunk@funknet.net> References: <5AC26DBC-7783-11D8-A204-00039317745E@functionalmedia.com><200403162300.12320.rfunk@funknet.net><4418.24.123.50.150.1079534395.squirrel@mail.jmglov.net> <200403171128.40626.rfunk@funknet.net> Message-ID: <4283.24.123.50.150.1079545907.squirrel@mail.jmglov.net> Quoth Rob Funk: > Josh Glover wrote: > >> With threads, you *don't* have to worry about open filehandles >> persisting through a fork (which can be a real issue when you have lots >> of filehandles open). > > I don't know what you're talking about here. Last I checked, threads share > filehandles. And fork doesn't work any differently from a threaded than > any other process. (Maybe you're thinking of Windows or some other > platform rather than Unix?) Nope. Here is an example of what I mean. Imagine that you have a process that is listening on a network socket for data, doing some transformation on it, and spewing it out on another network socket. Now imagine that when you encounter an error in the data stream, you need to send an email message to your pager so that you know that your attention is required. You don't want your main loop to have to worry about latency introduced by the interaction with the SMTP server, so you fork a child and use it to deal with the SMTP server. I have *seen*, in two separate and unrelated cases, the SMTP server interaction mess up the main process because of filehandles persisting through the fork. Closing the filehandles explicitly fixes the problem. Threads *share* filehandles, childrem *copy* them. There is an important distinction here. >> I think that the new generation of programmers, at least those with >> Computer Science degrees, have just as much knowledge about >> multi-threaded programming and the issues that surround it as they do >> about fork / exec-style programming. > > Probably more, in fact, which explains some of your misconceptions. :-) I don't have misconceptions, I just have problems communicating my opinions, apparently. I am familiar with both threaded and non-threaded programming in Unix. In fact, in the past six months, I have written one program in each style. > And this is mostly because those of us who took programming classes ten > years ago mostly learned on Unix, while those taking programming classes > now more often learn on Windows or Java. On those platforms, threading is > the way to go since forking a process is more expensive and works > differently from how it does in Unix. My programming courses were all in C++ at the beginning. In some of the upper-level classes that I took (Graphics, Compiler Construction, Networking), the professors prefered to read Java, so we used Java. Of course, my systems programming courses (Operating Systems, Unix Systems Programming, Linux Kernel Internals) were all in C. I haven't done any Windows programming since high school, and I am a happier person for it. :) >> The bottom line is, keep complexity down. For me, threads are less >> complex. For you, this might not be the case. We should both code >> accordingly. > > My first inclination is to agree with this, but then I remember that at > some point I'll have to maintain code written by you or someone like > you. :-) Given the number of multi-threaded programs that are written these days, I would say that most professional programmers should acquaint themselves with this style of programming. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 From pstjohn at hrblock.com Wed Mar 17 13:16:27 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Message Flag Message-ID: Thanks, I had just gotten to that by saving the mail as a file, and by getting a friend to extract stuff in his mail client. Presumably actukally breaking Outlook defeats the purpose of proslytizing in email? I can't read the witty, ironical jab at Microsoft; so really Jonadab is just preaching to the faithful. Peace comes from luring with commerce, not pushing with hostility. I want companies to drop MS because of the total costs of noncompetition and the rapid development in open-software, not because we throw mudballs at each other. Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Jeffrey Tadlock > Sent: Wednesday, March 17, 2004 11:54 AM > To: Central OH Linux User Group > Subject: Re: [COLUG] Message Flag > > On Wed, Mar 17, 2004 at 09:51:11AM -0600, St. John, Peter wrote: > > In MS Outlook, I can't read the mail from "Jonadab the Unsightly" apparently because "the message flag must be 100 bytes or less". > > > > It seems you are getting this because Jonadab has placed an > X-message-flag: in his mail headers that is too long for Outlook > to process. The message in his header is this: > > "This message is enhanced with advanced features that this version > of Outlook does not support. Please upgrade to the latest > version." > > X-message-flag is processed by the Outlook mail client and is a > "fun" way to flag an email message and give Outlook users a > "special" message. In the corporate world usually used by users > as the Follow-Up flag in Outlook to remind the recipient to > follow up on an action or email. > > Of course it can also be used to suggest to Outlook users to > upgrade their mail client to Mozilla, or your client of choice... > > > Is there a convenient workaround? > > Not sure there is a work around short of not using Outlook since > Outlook seems to only handle 100bytes in the field in question. > > /jft > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040317/6372b1b1/attachment.htm From sjs at khadrin.com Wed Mar 17 14:55:22 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Apache 1.3.29 vs Apache 2.0.48 In-Reply-To: <4054802A.2040201@skippy.net> References: <4054802A.2040201@skippy.net> Message-ID: <1079553322.19706.40.camel@cobra.khadrin.com> On Sun, 2004-03-14 at 10:54, Scott Merrill wrote: > The question is: do I stick with Apache 1.3.29 or plunge into > Apache2 2.0.48? There is some anecdotal evidence that Apache2 is not that popular yet. The following information was obtained by querying the Netcraft database at www.netcraft.com. [sjs@cobra netcraft]$ cat sites |xargs ./netcraft kernel.org: Apache/2.0.40 (Red Hat Linux) slashdot.org: Apache/1.3.29 (Unix) mod_gzip/1.3.26.1a mod_perl/1.29 sourceforge.net: Apache/1.3.27 (Unix) PHP/4.3.2 mod_ssl/2.8.12 OpenSSL/0.9.6b linux.com: Apache/1.3.29 (Unix) mod_gzip/1.3.26.1a mod_perl/1.28 newsforge.org: Apache/1.3.29 (Unix) mod_gzip/1.3.26.1a mod_perl/1.28 lwn.net: Apache linux-mag.com: Apache/1.3.12 (Unix) PHP/4.2.3 mod_ssl/2.6.6 OpenSSL/0.9.4 PHP/3.0.16 linuxjournal.com: Apache debian.org: Apache/1.3.26 (Unix) Debian GNU/Linux PHP/4.1.2 DAV/1.0.3 redhat.com: Apache www.gentoo.org: Apache/1.3.29 (Unix) (Gentoo/Linux) mod_gzip/1.3.26.1a AxKit/1.61 mod_perl/1.27 distrowatch.org: Apache/1.3.26 (Unix) Debian GNU/Linux PHP/4.3.4 mod_perl/1.26 kuro5hin.org: Apache/1.3.23 (Unix) mod_gzip/1.3.19.1a mod_ssl/2.8.6 OpenSSL/0.9.6a lkml.org: Apache/1.3.27 (Unix) Debian GNU/Linux mod_gzip/1.3.26.1a PHP/4.1.2 theregister.com: Apache/1.3.26 (Unix) Debian GNU/Linux news.com.com: Apache/2.0 apache.org: Apache/2.0.49-dev (Unix) netcraft.com: Apache/1.3.26 (Unix) mod_perl/1.27 Only three of the sites in this unscientific sampling were running Apache2. I remembered that kernel.org made the move to Apache2 in December, 2003. I figured a lot of other sites were running Apache2 by now as well. I was a bit disappointed. I like Apache2, but that is probably due to my (all too human) love of new toys. Below is the netcraft script. The lwp-request script comes with the LWP perl modules. [sjs@cobra netcraft]$ cat netcraft #!/bin/sh for site in $@ do lwp-request -m GET http://uptime.netcraft.com/up/graph?site=$site \ | perl -ne 'print "$1: " if /The site (.*)<\/a>/; print "$1\n" if /(.*)<\/b> on/' done exit From nliu99 at yahoo.com Wed Mar 17 16:12:36 2004 From: nliu99 at yahoo.com (Nan Liu) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] PXE Boot Problem In-Reply-To: <200403171554.i2HFs3FN001166@stones.wcbe.org> Message-ID: <20040317211236.98867.qmail@web80601.mail.yahoo.com> Hi! I have a compaq armada m300 that does not have a floppy drive or cdrom. The computer is ok since it boots into win2k, and the previous owner abandoned it since he couldn't load the intel nic driver onto the computer, so it was deemed "worthless" (bad battery too). Anyhow, I was planning to install linux on the laptop via PXE boot supported by the intel nic card. I was able to setup a dhcp/tftp server with bootp with PXELinux. I'm able to get to the boot: prompt with PXELinux loader. I attempted to load gentoo kernel and encountered: ... VFS: Cannot open root device "" or 03:03 Please append a correct "root=" boot option Kernel panic: VFS: unable to mount root fs on 03:03 The online howto regarding Gentoo is out of date, since it's for version 1.4. The howto on gentoo.org regarding alternative boot requires tinkering with grub, a nfs server, and a recompiled kernel. I suppose I could do that, but that's a lot of extra work just to do a one time OS install. Can anyone point me to a file that I can download to boot via PXElinux on the laptop to do a network installation of linux (prefer gentoo)? I tried the redhat pcmcia.img file, but PXElinux rejected it as invalid. Any hints? Thanks, Nan Liu From jep200305 at columbus.rr.com Wed Mar 17 16:36:12 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] PXE Boot Problem In-Reply-To: <20040317211236.98867.qmail@web80601.mail.yahoo.com> References: <200403171554.i2HFs3FN001166@stones.wcbe.org> <20040317211236.98867.qmail@web80601.mail.yahoo.com> Message-ID: <20040317163612.388ec1e7.jep200305@columbus.rr.com> Nan Liu wrote: > I have a compaq armada m300 that does not have a floppy drive or cdrom. > Can anyone point me to a file that I can download to boot via PXElinux > on the laptop to do a network installation of linux (prefer gentoo)? I > tried the redhat pcmcia.img file, but PXElinux rejected it as invalid. > Any hints? To test the PXE stuff, try the LTSP stuff under Knoppix (aka "Start Knoppix Terminal Server"). This won't install anything, and would only verify the possibility of using PXE (and Knoppix compatibility). From tom at functionalmedia.com Thu Mar 18 01:43:56 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] zope environment question Message-ID: I have the following ZPT or zope templating language snippet. I wrapped it in "pre" tags in case some mail program attempts to render the html. If the filters catch this I apologize. It seemed like a few html mails have gotten through lately so I thought I would try. This snippet takes the subfolders of a folder and builds links out of them. This works for the top level directory. I would like to call it from a subdirectory yet still present the subdirectories as viewed from the top. I want the same navbar for all folders with no difference based on level. How do I do this ? I tried placing this snippet in the top folder and calling it from one level down. It returned the values for the subdirectory. Any tips. Acquisition is a two edged sword. I want to run the script without it inheriting the environment of the resource that called it. I imagine that I can call it with a specific pathname or specify the path in the script but I am at a loss. This is my first day with zope templates. Any ideas ? Suppose I call this snippet "navbar" in my relatively top directory. Can I call it with a path such as here/navbar or home/navbar in order for it to get the right info ?? I plan on moving this whole site so absolute paths might not be the most flexible solution. I basically want to run a script that acts on the folder that it is located in no matter where it was called from. 

TITLE OR ID
Thanks Tom From John.Cramer at das.state.oh.us Thu Mar 18 14:45:17 2004 From: John.Cramer at das.state.oh.us (John.Cramer@das.state.oh.us) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Novell Launches New Version of SuSe Linux Message-ID: John.Cramer@das.state.oh.us on 03/18/2004 02:31:57 PM To: networking@ccscmh.org Subject: ccs/Novell Launches New Version of SuSe Linux Novell Launches New Version of SuSe Linux SEATTLE (Reuters) - Software company Novell Inc. unveiled on Thursday a new version of the Linux (news - web sites) operating system offered by a newly acquired unit, SuSe Linux AG, that aims to improve performance and the ability to run several programs at once. Novell (NasdaqNM:NOVL - news) said SuSe Linux 9.1, available in personal and professional editions, was the first to take advantage of the recently released core upgrade to Linux, the 2.6 kernel. All software programs, including Microsoft Corp.'s (NasdaqNM:MSFT - news) Windows, are based on a kernel that allows various programs to work on microprocessors. . . . Novell's Ungashick said installing the new Linux kernel would boost performance of existing PC hardware, enhancing the attractiveness of Linux. "This is a product that offers end users a choice," he said. Novell, a software maker known for its pioneering networking software, bought Germany's SuSe Linux for $210 million in January to launch its push into the market for Linux update and support services. == SuSE Linux 9.1 Springs from 2.6 Kernel Peter Galli - eWEEK SuSE Linux (news - web sites), now part of Novell Inc., on Thursday will announce that its SuSE Linux 9.1 personal and professional editions will be available in early May. The company says these are also the first complete commercial Linux distributions to support the recently released Linux 2.6 kernel. The update is also the first Linux distribution from the company since SuSE Linux AG and Ximian Inc. were acquired by Novell last year. This update is technically a point release, which are usually fairly minor upgrades. However, Charlie Ungershick, director of product management for the SuSE Linux business unit in Waltham, Mass., told eWEEK that this update is far more significant than a customary point release. . . . The updated SuSE Linux software will also support the GNOME 2.4.2 desktop, and there are several enhancements to this as well, including Windows-based printing configuration. Ungershick also reiterated that, unlike Red Hat Inc. and other major Linux players, Novell and SuSE remain committed to a Linux consumer desktop offering going forward. == http://www.novell.com/news/press/archive/2004/03/pr04018.html Included Applications SUSE LINUX 9.1 Personal and Professional continue to contain a wide range of applications. Standard pieces includes Internet, e-mail, image, graphics, audio and video software. In addition to text processing, spreadsheet and drawing applications, the office package OpenOffice.org 1.1 offers a formula editor and an HTML editor. OpenOffice.org exports PDF files with a mouse click, supports the import and export of various XML formats, and exports presentations in flash format. SUSE LINUX 9.1 is the first Linux package containing demo versions of the text processing application Textmaker and the spreadsheet application Planmaker from Softmaker. The database application Rekall and the latest version of the home banking software Moneyplex* from Matrica are also new. Professional layouts can be prepared with the desktop publishing application Scribus. SUSE LINUX Personal ? Simple Migration, No Installation Needed SUSE LINUX 9.1 Personal is the ideal solution for customers who want to use Linux exclusively on the desktop. The new LiveCD enables Linux newcomers to get to know and test the functions and applications of the new system without installing it on the hard disk. A complete user manual helps newcomers get acquainted with the world of Linux. The second CD can be used to install SUSE LINUX Personal and the integrated firewall in a few easy steps, either as the only operating system or alongside any Windows operating system (dual boot). YaST Online Update makes sure the system is always up to date and secure. Available in Early May SUSE LINUX 9.1 will be available at http://store.suse.com and from bookstores and software suppliers on May 6. The recommended retail price of SUSE LINUX 9.1 Personal (two CDs, installation guide, 30 days of installation support) is $29.95. SUSE LINUX 9.1 Professional (five CDs, two double-sided DVDs, user guide and administration guide, 90 days of installation support) is $89.95. The update edition of SUSE LINUX 9.1 Professional is $59.95. == Full announcements are at these (links) web sites: Novell Launches New Version of SuSe Linux http://story.news.yahoo.com/news?tmpl=story&u=/nm/20040318/tc_nm/tech_linux_suse_dc SuSE Linux 9.1 Springs from 2.6 Kernel http://news.yahoo.com/news?tmpl=story2&cid=1738&u=/zd/20040317/tc_zd/121906&printer=1 http://www.novell.com/news/press/archive/2004/03/pr04018.html == BCC: CCSLinux, CCSnetworking & linux-sig@dma == JohnW Cramer === JWCramer-HillTopUSA jcramer_css@hotmail.com Columbus Peregrine Falcon Daylight Cam: http://www.dnr.state.oh.us/wildlife/falcons/default.htm "Genealogy without documentation is mythology." Maggie Ohio Ancestors: CRAMER MARCHAL SCHMITTHIPPLE TRICK LOY From linux at litenverden.org Thu Mar 18 17:09:08 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Privoxy Question In-Reply-To: References: Message-ID: <405A1E04.7010803@litenverden.org> I have a Privoxy question. I hope we have an expert somewhere. I want to block any connection to: *ru4.com/* Here is an actual example URL: http://http300.edge.ru4.com/smartserve/click?placement=pl-0363-912&opt=1079646950|363|ad-0363-251|pl-0363-912|control|0|pl-0363-912%26mid+atlantic%26afternoon%26noinfo%26high%260%26U1|28&pr=&epr=&dmurl=&edmurl=&target=http://free.aol.com/tryaolfree/index.adp?promo=472562&service=aolhsb Here is what I put into /etc/privoxy/user.action: { +block } www.example.com/nasty-ads/sponsor.gif # Here we found one that is not in Privoxy's default blocked list: #.adbureau.net *.ru4.com/* *.doubleclick.net/* This seems to have no effect. Thoughts? George From whooper at freeshell.org Thu Mar 18 18:47:13 2004 From: whooper at freeshell.org (William Hooper) Date: Sat Jan 8 01:34:46 2005 Subject: [COLUG] Novell Launches New Version of SuSe Linux In-Reply-To: References: Message-ID: <65388.65.41.50.216.1079653633.squirrel@65.41.50.216> John.Cramer@das.state.oh.us said: > Available in Early May SUSE LINUX 9.1 will be available at > http://store.suse.com and from bookstores and software suppliers on May 6. If this isn't just a ploy to be "first" I don't know what is. -- William Hooper From mchenrytech at wideopenwest.com Fri Mar 19 01:48:30 2004 From: mchenrytech at wideopenwest.com (Sean McHenry) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Off topic - Amateur Radio Related Message-ID: <001b01c40d7e$30de9370$6401a8c0@bigedit> Just in case some of the fellow Amateur Radio Operators are interested, I am selling a Drake TR4C on Ebay. Not Linux related but some of us are operators so I thought it might be of general interest to a portion of the group. See details at my personal web site. http://65.60.189.39/Ebay.html That's it, Thanks for the bandwidth. Sean McHenry Sean@McHenryTech.com From jep200305 at columbus.rr.com Fri Mar 19 09:22:28 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Yast going GPL Message-ID: <20040319092228.0894a0e8.jep200305@columbus.rr.com> Apparently, Yast is going GPL: http://news.com.com/2100-7344_3-5175682.html, so maybe Novell "gets it". From jonadab at bright.net Fri Mar 19 11:27:16 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Message Flag In-Reply-To: References: Message-ID: "St. John, Peter" writes: > Presumably actukally breaking Outlook defeats the purpose of > proslytizing in email? I can't read the witty, ironical jab at > Microsoft; so really Jonadab is just preaching to the faithful. I didn't know it was breaking, because I didn't know about the 100-byte limit. It should be fixed now. I've never had anyone be unable to read the messages before, so far as I'm aware, and I put that message flag in quite some while ago when I was messing around with custom headers (about the same time I added the X-Face header IIRC), so it seems likely that at least some versions of Outlook or Outlook Express do not have this problem. Hard to be sure, though. > Peace comes from luring with commerce, not pushing with hostility. Wasn't intended to be hostile, just a good-natured jab. > I want companies to drop MS because of the total costs of > noncompetition and the rapid development in open-software, not > because we throw mudballs at each other. I don't care if anybody drops Microsoft as a vendor in general or not (I use Windows myself, though not as my primary system), but I want people to stop using Outlook because I'm sick of giving up a third of my bandwidth for days at a time to the latest new Outlook virus/worm/trojan every month or two. Outlook is the one and only piece of Microsoft software that I seriously wish had never been created. But it was not my intention to take this out on users by making the messages actually unreadable; I didn't know Outlook was *that* broken. It was only supposed to show a message flag. I've shortened the flag to less than 100 bytes now. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Fri Mar 19 19:03:59 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Yast going GPL In-Reply-To: <20040319092228.0894a0e8.jep200305@columbus.rr.com> References: <20040319092228.0894a0e8.jep200305@columbus.rr.com> Message-ID: Jim P writes: > Apparently, Yast is going GPL If this had happened three years ago, some other distributions might have adopted it, but by now they all have their own equivalent, so I'm guessing that won't happen. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From linux at litenverden.org Fri Mar 19 19:28:56 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Referrers In-Reply-To: <405239E6.7030402@litenverden.org> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> <405239E6.7030402@litenverden.org> Message-ID: <405B9048.3020609@litenverden.org> Does anyone know how can I disable referrers with Mozilla under Linux? I'm also running Privoxy. On my Win2k box, I use Mozilla and CookieCop 2 instead of Privoxy. CookieCop has a referrer "switch" that I can flip to disable. I just realized that Moz/Linux was sending referrers. Bad thing. George From sjs at khadrin.com Fri Mar 19 19:42:26 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Referrers In-Reply-To: <405B9048.3020609@litenverden.org> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> <405B9048.3020609@litenverden.org> Message-ID: <1079743345.29335.3.camel@cobra.khadrin.com> On Fri, 2004-03-19 at 19:28, George H. Yeager wrote: > Does anyone know how can I disable referrers with Mozilla under Linux? > I'm also running Privoxy. Hi. You can access a bunch of config options by typing "about:config" into the location bar. If that page is searchable then I don't know how. I think the option you are interested in might be "network.http.sendRefererHeader". Good luck. -- Stephen J. Smith | sjs@khadrin.com | http://khadrin.com/ From phostetl at columbus.rr.com Fri Mar 19 19:59:41 2004 From: phostetl at columbus.rr.com (Paul Hostetler) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Yast going GPL In-Reply-To: References: <20040319092228.0894a0e8.jep200305@columbus.rr.com> Message-ID: On Mar 19, 2004, at 7:03 PM, Jonadab the Unsightly One wrote: > Jim P writes: > >> Apparently, Yast is going GPL > > If this had happened three years ago, some other distributions might > have adopted it, but by now they all have their own equivalent, so I'm > guessing that won't happen. I'll bite. Which distribution has something equivalent to Yast? Yast doesn't suck. From linux at litenverden.org Fri Mar 19 20:23:30 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Referrers In-Reply-To: <1079743345.29335.3.camel@cobra.khadrin.com> References: <20040127142442.52C8B304021@dalamar.ilive4code.net> <405232DB.6020302@litenverden.org> <405B9048.3020609@litenverden.org> <1079743345.29335.3.camel@cobra.khadrin.com> Message-ID: <405B9D12.9080500@litenverden.org> Stephen J. Smith wrote: > On Fri, 2004-03-19 at 19:28, George H. Yeager wrote: > >>Does anyone know how can I disable referrers with Mozilla under Linux? >>I'm also running Privoxy. > > > Hi. You can access a bunch of config options by typing "about:config" > into the location bar. If that page is searchable then I don't know > how. I think the option you are interested in might be > "network.http.sendRefererHeader". Good luck. > I found the config change I need. It is in the default.action file. I made the change carefully (I'm sure it's correct), and now Privoxy is refusing connections to the browser. It won't work when I put things back, either. Guess I'll remove Privoxy and reinstall. George From jep200305 at columbus.rr.com Fri Mar 19 20:49:30 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] AC Powerline Quality In-Reply-To: <402D5543.4040905@bourbaki.us> References: <20040212155434.393f26c1.jep200305@columbus.rr.com> <1076626988.2123.1.camel@linux.local> <20040213202601.GA15060@quillandmouse.com> <20040213161759.7aa2839d.jep200305@columbus.rr.com> <00b601c3f27a$cb9b7280$6501a8c0@wowway.com> <402D5543.4040905@bourbaki.us> Message-ID: <20040319204930.7fdf3964.jep200305@columbus.rr.com> Everything you wanted to know about Watts versus VA and more: http://www.reed-electronics.com/ednmag/toc/3/18/2004 It also mentions how non-linear loads, such as typical computers, can cause unexpected resets. From linux at litenverden.org Fri Mar 19 21:00:00 2004 From: linux at litenverden.org (George H. Yeager) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Referrers In-Reply-To: <5789A42DD684734298BF9EC852D35111033AB018@scad1exis02.alldata.net> References: <5789A42DD684734298BF9EC852D35111033AB018@scad1exis02.alldata.net> Message-ID: <405BA5A0.8060100@litenverden.org> OK, I got it. The best thing to do is to edit the action files with the web-based tool built into Privoxy . . .duh. I set it up to forge the referrer to the root of the site I'm visiting. Works just fine. George From n2uro at aol.com Fri Mar 19 21:00:27 2004 From: n2uro at aol.com (Samuel Greenfeld) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Ohio Linuxfest (Ohio State University, Oct. 2nd) - Call for Presentations, Attendees Message-ID: <405BA5BB.50301@aol.com> The Ohio LinuxFest is a small opensource software convention held in Ohio. Last year marked its kickoff, with approximately 100 attendees present. This year, the event will ideally again be held at the Ohio State University Columbus campus, although we would like to increase the event in size. Provided rooms are available, the event will be held Saturday, October 2, 2004 (no local football game). If you would be interested in attending, or possibly presenting a piece about your favorite opensource security program (OpenBSD, Snort, LIDS, SELinux, etc.) or usage thereof, please look at www.ohiolinux.org and/or follow the instructions below. Sincerely, Samuel Greenfeld -------- Original Message -------- Subject: [InterLUG] Ohio LinuxFest 2004 Announcement Date: Wed, 10 Mar 2004 22:10:19 -0500 From: Michael Meffie To: interlug@cantonlinux.org Ohio LinuxFest 2004 October 2, 2004 Ohio State University, Columbus, Ohio http://www.ohiolinux.org Call for Presentations Please send the following information to share@ohiolinux.org by May 15, 2004: * Title of Presentation * Short Description * Short Bio * Intended Audience (Technical, Non-Technical) * Preferred Format & Time Allocation * Required A/V & Other Equipment * Any Special Requirements Linux User Groups * A table will be provided for each LUG that would like to participate. * Please contact me if your LUG would like to participate. T-Shirt Design Contest * Please send your submissions by May 15, 2004 * See http://ohiolinux.org/tshirt.html for details We welcome any comments, suggestions, or questions. Send email to team@ohiolinux.org. Please feel free to forward this announcement. From jep200305 at columbus.rr.com Fri Mar 19 21:38:10 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Thousands of eyeballs aren't always a good thing Message-ID: <20040319213810.65a5b770.jep200305@columbus.rr.com> Here's the bait: http://www.reed-electronics.com/ednmag/toc/3/18/2004 From jep200305 at columbus.rr.com Fri Mar 19 21:43:12 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] AC Powerline Quality In-Reply-To: <20040319204930.7fdf3964.jep200305@columbus.rr.com> References: <20040212155434.393f26c1.jep200305@columbus.rr.com> <1076626988.2123.1.camel@linux.local> <20040213202601.GA15060@quillandmouse.com> <20040213161759.7aa2839d.jep200305@columbus.rr.com> <00b601c3f27a$cb9b7280$6501a8c0@wowway.com> <402D5543.4040905@bourbaki.us> <20040319204930.7fdf3964.jep200305@columbus.rr.com> Message-ID: <20040319214312.4438cb4f.jep200305@columbus.rr.com> Earlier I wrote: > http://www.reed-electronics.com/ednmag/toc/3/18/2004 I meant: http://www.reed-electronics.com/ednmag/article/CA402134?pubdate=3%2F18%2F2004 From jep200305 at columbus.rr.com Fri Mar 19 21:45:15 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Thousands of eyeballs aren't always a good thing In-Reply-To: <20040319213810.65a5b770.jep200305@columbus.rr.com> References: <20040319213810.65a5b770.jep200305@columbus.rr.com> Message-ID: <20040319214515.64b232b5.jep200305@columbus.rr.com> Earlier I wrote: > Here's the bait: > > http://www.reed-electronics.com/ednmag/toc/3/18/2004 I meant: http://www.reed-electronics.com/ednmag/article/CA402140?pubdate=3%2F18%2F2004 From blata at extent0006.entomology.ohio-state.edu Fri Mar 19 23:04:25 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] utf-8 Message-ID: <405BC2C9.3020701@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've been having trouble with one of my log files. When I try to view it with less or more I get this Log: Log file open, Fri Mar 19 22:49:34 2004 Init: Name subsystem initialized L^@o^@g^@:^@ ^@Y^@o^@u^@r^@ ^@l^@o^@c^@a^@l^@e^@:^@ ^@[^@U^@T^@F^@-^@8^@]^@.^@ ^@ ^@L^@o^@g^@:^@ ^@E^@n^@a^@b^@l^@e^@d^@ ^@U^@N^@I^@C^@O^@D^@E^@ ^@s^@u^@p^@p^@o^@r^@t^@.^@ ^@I^@n^@i^@t^@:^@ ^@V^@e^@r^@s^@i^@o^@n^@:^@ ^@3^@1^@8^@6^@ ^@(^@1^@2^@7^@.^@2^@9^@)^@ ^@I^@n^@i^@t When I try to use gedit I get this: Could not open the file "/home/blata/.ut2004/System/UT2004.log" because it contains invalid UTF-8 data. what does this mean and how do I fix it? - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAW8LJv+6+qSFv3TURAhV+AJ0SxMzIRY36/igUDwn5RhyMGLtRJACdEU+B QpN4TgSm8qyMawOSANpl3QU= =nYe8 -----END PGP SIGNATURE----- From rfunk at funknet.net Sat Mar 20 01:52:43 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] utf-8 In-Reply-To: <405BC2C9.3020701@bugs.osu.edu> References: <405BC2C9.3020701@bugs.osu.edu> Message-ID: <200403200152.43888.rfunk@funknet.net> Wade Pinkston wrote: > I've been having trouble with one of my log files. > When I try to view it with less or more I get this > > Log: Log file open, Fri Mar 19 22:49:34 2004 > Init: Name subsystem initialized > L^@o^@g^@:^@ ^@Y^@o^@u^@r^@ ^@l^@o^@c^@a^@l^@e^@:^@ > Could not open the file "/home/blata/.ut2004/System/UT2004.log" > because it contains invalid UTF-8 data. Looks like gedit recognizes (somehow) that it's Unicode, but doesn't realize that it's utf-16 or ucs-2 rather than utf-8. The key is that every other byte appears to be a zero byte (which is printed as ^@). UCS-2 is a straight two-byte encoding. I think it ignores characters that don't fit in two bytes. UTF-8 is the same as ASCII for the 0-127 ASCII characters; for characters 128-255 it's almost like iso-8859-1 but has an extra byte inserted per character. For higher characters you get more bytes inserted. UTF-16, as I understand it, sorta combines the two previous ones -- it's basically a two-byte format, but inserts extra bytes for characters that don't fit into two bytes. "man 7 unicode" might help explain things more thoroughly. "man -k unicode" might tell you about programs on your system that work with unicode. The simplest way to translate that file into something readable, assuming it doesn't use any unicode characters outside the ASCII range, might be this: tr -d '\000' < /home/blata/.ut2004/System/UT2004.log | more -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jmccluggage at neo.rr.com Sat Mar 20 15:39:27 2004 From: jmccluggage at neo.rr.com (Jody McCluggage) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Linux OS that runs off a cd In-Reply-To: <200402292351.58732.rfunk@funknet.net> Message-ID: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Hello, I want to build a Linux system optimized for running security tools that will run from a cd . Does anyone know of any online articles that walk you through creating a linux system that runs off a cd? Thanks, Jody --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.592 / Virus Database: 375 - Release Date: 2/18/2004 From george at byrd.net Sat Mar 20 15:49:05 2004 From: george at byrd.net (George Byrd) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Linux OS that runs off a cd In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <405CAE41.6020403@byrd.net> Jody, There was an article on http://www.slashdot.org/ a short while back that listed an attempt to catalog all linux-on-CD distributions. There were quite a few, possibly over 100, and this included distributions with security tools already built in. Instead of re-inventing the wheel, you might try searching that list. I'll take a poke around myself and see if I can find you a better link, but in the meantime, feel free to search Slashdot or Google. Good luck! George :) Jody McCluggage wrote: >Hello, > >I want to build a Linux system optimized for running security tools that >will run from a cd . Does anyone know of any online articles that walk you >through creating a linux system that runs off a cd? > >Thanks, > >Jody > >--- >Outgoing mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.592 / Virus Database: 375 - Release Date: 2/18/2004 > > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > > > > From phillip_the_ambidexterous at yahoo.com Sat Mar 20 16:53:14 2004 From: phillip_the_ambidexterous at yahoo.com (Phillip the Ambidexterous) Date: Sat Jan 8 01:34:47 2005 Subject: [COLUG] Linux OS that runs off a cd In-Reply-To: <405CAE41.6020403@byrd.net> Message-ID: <20040320215315.12724.qmail@web60808.mail.yahoo.com> --- George Byrd wrote: > Jody, > > There was an article on http://www.slashdot.org/ a > short while back that > listed an attempt to catalog all linux-on-CD > distributions. There were > quite a few, possibly over 100, and this included > distributions with > security tools already built in. ... That would be: http://www.frozentech.com/content/livecd.php -Phil __________________________________ Do you Yahoo!? Yahoo! Finance Tax Center - File online. File on time. http://taxes.yahoo.com/filing.html From jep200305 at columbus.rr.com Sat Mar 20 17:50:00 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Linux OS that runs off a cd In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200402292351.58732.rfunk@funknet.net> <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <20040320175000.591a6a76.jep200305@columbus.rr.com> Jody McCluggage wrote: > I want to build a Linux system optimized for running security tools that > will run from a cd . Several security live CDs have been built already. > Does anyone know of any online articles that walk you > through creating a linux system that runs off a cd? Some of the live CD Linuxes make it easy for you to do so. You install them to the hard drive, make whatever changes you want, and then run the magic command to make a new bootable live CD image. Look through http://www.distrowatch.com/dwres.php?resource=cd for the versions such as PCLinuxOS that has the mklivecd command. http://www.distrowatch.com/table.php?distribution=pclinuxos That should be easier than reading an article. From jep200305 at columbus.rr.com Sat Mar 20 17:55:26 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Making one's own Linux live CD In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200402292351.58732.rfunk@funknet.net> <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <20040320175526.5885126a.jep200305@columbus.rr.com> Jody McCluggage wrote: > I want to build a Linux system optimized for running security tools that > will run from a cd . Does anyone know of any online articles that walk you > through creating a linux system that runs off a cd? http://www.distrowatch.com/table.php?distribution=eagle says: It also offers an easier way for less experienced Linux users to create their own bootable floppy or CD from scratch using the HOW-TOs available on the downloads page. http://www.distrowatch.com/dwres.php?resource=cd says: The project differs from the rest of the distributions listed on this page in that Eagle Linux is not a complete, ready-to-use Linux CD, but rather a document describing how to create one. As such, it is a great learning tool, suitable for educational establishments. ... version 1.0 builds a bootable CD from of an existing Red Hat installation, while version 2.0 does the same for Debian users. Besides building a CD, the documents can also be adapted to building embedded Linux systems running on various platforms and processors From jep200305 at columbus.rr.com Sat Mar 20 18:01:53 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Making one's own Linux live CD: article In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200402292351.58732.rfunk@funknet.net> <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <20040320180153.105fdf96.jep200305@columbus.rr.com> Jody McCluggage wrote: > I want to build a Linux system optimized for running security tools that > will run from a cd . Does anyone know of any online articles that walk you > through creating a linux system that runs off a cd? http://nwst.de/livelinuxcd/howto/toolkit-quickstart/t1.html http://www.distrowatch.com/table.php?distribution=rpmlive From jerrylowery at wideopenwest.com Sat Mar 20 18:30:51 2004 From: jerrylowery at wideopenwest.com (Jerry Lowery) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Linux OS that runs off a cd In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <405CD42B.9000801@wideopenwest.com> Devil Linux. Jerry Jody McCluggage wrote: >Hello, > >I want to build a Linux system optimized for running security tools that >will run from a cd . Does anyone know of any online articles that walk you >through creating a linux system that runs off a cd? > >Thanks, > >Jody > > > From mwmiller at columbus.rr.com Sat Mar 20 18:24:05 2004 From: mwmiller at columbus.rr.com (Matthew W. Miller) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Yast going GPL In-Reply-To: References: <20040319092228.0894a0e8.jep200305@columbus.rr.com> Message-ID: <20040320232405.GA2301@columbus.rr.com> On Fri, Mar 19, 2004 at 07:59:41PM -0500, Paul Hostetler wrote: >I'll bite. Which distribution has something equivalent to Yast? >Yast doesn't suck. Neither does the poldek: http://team.pld.org.pl/~mis/poldek/ And yes, it's been GPLv2 licensed all along. -- Matthew W. Miller MPlayer does not suck. From sjs at khadrin.com Sat Mar 20 18:41:24 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Yast going GPL In-Reply-To: <20040320232405.GA2301@columbus.rr.com> References: <20040319092228.0894a0e8.jep200305@columbus.rr.com> <20040320232405.GA2301@columbus.rr.com> Message-ID: <1079826083.6863.6.camel@cobra> On Sat, 2004-03-20 at 18:24, Matthew W. Miller wrote: > On Fri, Mar 19, 2004 at 07:59:41PM -0500, Paul Hostetler wrote: > >I'll bite. Which distribution has something equivalent to Yast? > >Yast doesn't suck. > > Neither does the poldek: http://team.pld.org.pl/~mis/poldek/ And yes, > it's been GPLv2 licensed all along. YaST is much more than a dependency-resolving update program. It also let's you configure just about everything on your system. Some people like that and some people hate it. There is probably no convincing either side. From hcrouch at mchsi.com Sat Mar 20 20:28:04 2004 From: hcrouch at mchsi.com (Harold Crouch) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Those Pesky Module Paths Message-ID: <20040320192804.1b88d4ec.hcrouch@mchsi.com> All of the following occurred in an x-term session. I downloaded, untarred, configured and made (but did *not* make-install) the XFce3 source file. I did this to obtain one specific application: xftree. As an unpriviliged user, I can copy xftree from /usr/src/xfce3/xftree/ into any folder on my computer, do "./xftree," and xftree launches. As root, I can copy xftree into any folder on my computer, do "./xftree," and xftree launches *and* returns the warning message "Unable to locate loadable module in module path: 'libxfce.so.'" The module does exist in one of the folders under /usr/src/xfce3/. I tried putting it in the same folder as xftree, but xftree didn't see it. I tried putting xftree in /usr/local/bin/ *and* putting the module in /usr/local/lib, but still no joy. How do I determine what is the correct module path without doing a make-install? Is there a makefile that I can read through that will show me the target folders? Else, how does one pass module location to a binary in a BASH startup script? Thanks, - Harold From jep200305 at columbus.rr.com Sat Mar 20 22:05:24 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Those Pesky Permissions In-Reply-To: <20040320192804.1b88d4ec.hcrouch@mchsi.com> References: <20040320192804.1b88d4ec.hcrouch@mchsi.com> Message-ID: <20040320220524.3aa9d224.jep200305@columbus.rr.com> Harold Crouch wrote: > As an unpriviliged user, I can copy xftree from /usr/src/xfce3/xftree/ > into any folder on my computer, This sounds odd. It is unusual for an unpriviledged user to be able to copy something into _any_ folder. Typically, there are many folders that an unpriviledged user most definitely can _not_ copy stuff into. E.g., an unpriviledged user can _not_ copy stuff into /bin on most Linux computers. As an unpriviliged user, can you copy xftree to /bin? From hcrouch at mchsi.com Sat Mar 20 22:45:18 2004 From: hcrouch at mchsi.com (Harold Crouch) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Those Pesky Permissions In-Reply-To: <20040320220524.3aa9d224.jep200305@columbus.rr.com> References: <20040320192804.1b88d4ec.hcrouch@mchsi.com> <20040320220524.3aa9d224.jep200305@columbus.rr.com> Message-ID: <20040320214518.2be2b814.hcrouch@mchsi.com> HC>> As an unpriviliged user, I can copy xftree from HC>> /usr/src/xfce3/xftree/ into any folder on my computer, JP> This sounds odd. JP> JP> It is unusual for an unpriviledged user to be able to copy JP> something into _any_ folder. Typically, there are many folders JP> that an unpriviledged user most definitely can _not_ copy JP> stuff into. E.g., an unpriviledged user can _not_ copy JP> stuff into /bin on most Linux computers. JP> JP> As an unpriviliged user, can you copy xftree to /bin? My fingers were moving faster than my brain. It seems to be happening a lot lately. :-( Let's see if I can get it right this time. You are correct. I could *not* freely copy into */bin folders as an unpriviliged user, but once root had put the binary in those locations, the unpriviliged user was able to execute all of them. In retrospect, I'm not sure why that bothered me; unpriviliged users *have* to be able to execute binaries in the */bin folders. From jep200305 at columbus.rr.com Sat Mar 20 23:07:52 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Making one's own Linux live CD In-Reply-To: <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> References: <200402292351.58732.rfunk@funknet.net> <200403202039.i2KKdVtO025169@ms-smtp-02-eri0.ohiordc.rr.com> Message-ID: <20040320230752.1aa0d822.jep200305@columbus.rr.com> Jody McCluggage wrote: > Does anyone know of any online articles that walk you > through creating a linux system that runs off a cd? http://livecd.berlios.de/about.php From blitz at post891.org Sun Mar 21 07:59:00 2004 From: blitz at post891.org (Patrick Blitz) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Those Pesky Permissions In-Reply-To: <20040320214518.2be2b814.hcrouch@mchsi.com> References: <20040320192804.1b88d4ec.hcrouch@mchsi.com> <20040320220524.3aa9d224.jep200305@columbus.rr.com> <20040320214518.2be2b814.hcrouch@mchsi.com> Message-ID: <1079873941.657.2.camel@amon> On Sun, 2004-03-21 at 04:45, Harold Crouch wrote: > You are correct. I could *not* freely copy into */bin folders as an > unpriviliged user, but once root had put the binary in those > locations, the unpriviliged user was able to execute all of them. In > retrospect, I'm not sure why that bothered me; unpriviliged users > *have* to be able to execute binaries in the */bin folders. > I think what your Problem is that the Display Variables are not passed on. On my System, when I su to root an try to execute let's say gdmconfig, it doesn't find the X-server and quits with weird errors. When i execute gdmconfig using sudo, it runs. Try executing it using sudo. From jonadab at bright.net Sun Mar 21 23:10:37 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Yast going GPL In-Reply-To: References: <20040319092228.0894a0e8.jep200305@columbus.rr.com> Message-ID: <3c81r076.fsf@jonadab.homeip.net> Paul Hostetler writes: > I'll bite. Which distribution has something equivalent to Yast? Mandrake for example has a suite of tools that accomplishes basically the same things as Yast. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From skippy at skippy.net Mon Mar 22 08:20:38 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] March Meeting Announcement Message-ID: <32943.216.136.35.122.1079961638.squirrel@www.skippy.net> Followup-to: colugx at colug.net MEETING NOTICE ============== Central Ohio Linux Users Group Date: Wednesday, 31 March, 7:00 PM to 9:00 PM local Meeting Presentation ==================== Topic: Novell's GNU/Linux initiatives Mark Richards, Senior Product Specialist at Novell, will discuss Novell's GNU/Linux initiatives, including SuSE Linux, Ximian, and more. The upcoming meeting is open to the general public, and all with an interest in Free Software, Open Source, Apache, MySQL, Perl, PHP, postfix, Linux, BSD, Unix, Windows, or whatever, are invited. MEETING SITE: ============= The March 2004 meeting will be held at: OCSEA 390 Worthington Rd. Westerville, OH 43082 865-4700 ...The group has met on the LAST Wednesday or LAST Saturday of a month, alternating by month to allow more folks to mesh their schedules with COLUG's. Sites and dates also depend on facility availabity, with common-sense relocation in the holiday season. The webpage often has a link to a map to the upcoming meeting site. See: http://www.colug.net/meetings.php The website is always authoritative to other discussions, such as might be seen on the mailing list. From tom at functionalmedia.com Mon Mar 22 13:55:56 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] threads and multiple processors Message-ID: <8D5057FD-7C32-11D8-981A-00039317745E@functionalmedia.com> COLUG, I have one remaining question about threads. It seems that threads are most often mentioned when talking about multiple processors. Is a threaded program required to take advantage of multiple processors ? Did this issue have anything to do with the push for threads in Apache2 ?? Just curios. Thanks Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From tnoe at mailsnare.net Mon Mar 22 14:30:53 2004 From: tnoe at mailsnare.net (tnoe@mailsnare.net) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] threads and multiple processors Message-ID: Creating multi or free-threaded applications is normally considered as one of the original design decisions when creating a program. Some computing tasks lend themselves very well to a free-threaded model. Others do not. So, one could argue that most single-threaded apps can be written as multi-threaded and vice versa. And the number of processors isn't necessarily relevant. Using the wrong model can lead to a significantly slower runtime. Complex tasks that are easily broken up into smaller tasks are usually good candidates for multi-threaded applications. If the tasks also have no affinity towards a given processor, then they will also work well on multi-processor systems. So, the answer to your question is, multi-threaded applications that are designed not to have processor affinity should run faster than a similarly designed single-threaded application on a multi-threaded processor system. And single-threaded applications can still run on a multi-processor system. -- Best wishes, Tom E: tnoe@mailsnare.net P: tnoe@vtext.com >-----Original Message----- >From: tom hanlon [mailto:tom@functionalmedia.com] >Sent: Monday, March 22, 2004 06:55 PM >To: colug@colug.net >Subject: [COLUG] threads and multiple processors > >COLUG, > >I have one remaining question about threads. > >It seems that threads are most often mentioned when talking about >multiple processors. Is a threaded program required to take advantage >of multiple processors ? Did this issue have anything to do with the >push for threads in Apache2 ?? > >Just curios. > >Thanks > > > > > >Tom Hanlon >Functional Media >740-597-1472 >tom@functionalmedia.com >www.functionalmedia.com > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > From jep200305 at columbus.rr.com Mon Mar 22 17:44:04 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Linux DTP package.... In-Reply-To: <200103011931.OAA00299@big.seorf.ohiou.edu> References: <200103011931.OAA00299@big.seorf.ohiou.edu> Message-ID: <20040322174404.1f146390.jep200305@columbus.rr.com> Tom Hanlon wrote: > ... Chilliware Ice Sculptor Desktop ... How did this compare to Gimp? I get the impression that Chilliware has gone to that big bit bucket in the sky. From tom at functionalmedia.com Tue Mar 23 00:20:43 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Linux DTP package.... In-Reply-To: <20040322174404.1f146390.jep200305@columbus.rr.com> Message-ID: I did not post any post regarding chiliware ice sculptor Desktop. At least not intentionally. Interesting.. Tom On Monday, March 22, 2004, at 05:44 PM, Jim P wrote: > Tom Hanlon wrote: > >> ... Chilliware Ice Sculptor Desktop ... > > How did this compare to Gimp? > > I get the impression that Chilliware > has gone to that big bit bucket in the sky. > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From rfunk at funknet.net Tue Mar 23 01:16:27 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] threads and multiple processors In-Reply-To: <8D5057FD-7C32-11D8-981A-00039317745E@functionalmedia.com> References: <8D5057FD-7C32-11D8-981A-00039317745E@functionalmedia.com> Message-ID: <200403230116.27230.rfunk@funknet.net> tom hanlon wrote: > It seems that threads are most often mentioned when talking about > multiple processors. Is a threaded program required to take advantage > of multiple processors ? Not exactly, but they can help. There's one way to waste multiple CPUs: run a single process that doesn't fork or thread. DOS is good at this. There are three ways to take advantage of multiple CPUs. One is to run multiple individual processes that don't themselves fork or thread. The different processes may run simultaneously on different CPUs. For example, lots of different users running their own programs on a modern SMP Linux machine. Another is to run a process that forks itself into multiple processes. Again, the different processes may run on different CPUs. Apache 1 is a good example of this. The third is to run a process that splits into threads. The different threads may run on different CPUs. Apache 2 is an example of this. Event-loop-based servers such as thttpd fall somewhere between case zero (single process) and case 1 (multiple unrelated processes), since the server program itself doesn't take advantage of multiple CPUs but a Unix machine running one of these can still take some small advantage of multiple CPUs since it usually has other processes to run besides that one server program. Note that the "S" in SMP is "symmetric", meaning that all processors have equal access to memory. This is extremely useful for threads, which share their memory; if you have two threads running on two different CPUs, you want them both to have equivalent access to the memory they share. Separate unrelated processes (case 1) don't care about equal access to memory at all, except for shared libraries. Separate related processes (case 2) care only a little bit; they tend to look at the same memory until it gets written, at which time it gets split per-process so that each process has its own memory to write to. SMP can be contrasted with NUMA, where Memory Access is Non-Uniform between CPUs, so the OS needs to take more care with what CPUs run what threads. (It's been noted on LKML that an SMP machine with multiple hyperthreading-capable processors could/should be treated as a NUMA system, since the virtual processors within a single processor have access to shared cache memory that the other processors can't get to.) > Did this issue have anything to do with the > push for threads in Apache2 ?? No, since multiple processors split the multiple processes of Apache 1 just fine. I believe the main motivation of Apache 2 was to improve Windows performance. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Tue Mar 23 08:31:57 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] Chilliware Deja-Vu In-Reply-To: References: <20040322174404.1f146390.jep200305@columbus.rr.com> Message-ID: <20040323083157.17d50d5e.jep200305@columbus.rr.com> On Tue, 23 Mar 2004 00:20:43 -0500 tom hanlon wrote: > I did not post any post regarding chiliware ice sculptor Desktop. > At least not intentionally. It's been long enough ago that you just forgot. http://static.colug.net/cooked/v04.n082.html BTW, Chilliware went bye-bye about three years ago. From ken at alpha2.com Tue Mar 23 10:23:08 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:48 2005 Subject: [COLUG] modules.conf Message-ID: <002601c410ea$bebc0000$690aa8c0@alpha2.com> Can anyone tell me the correct way to move: rmmod ide-tape modprobe ide-scsi hdparm -d0 /dev/hdt0 mt -f /dev/st0 stopoptions no-blklimits mt -f /dev/st0 stopoptions no-blklimits from /etc/rc.d/rc.local to /etc/modules.conf? Ken Bradford Alpha II Service, Inc. From nordlus at ilive4code.net Tue Mar 23 10:47:22 2004 From: nordlus at ilive4code.net (Stephen Nordlund) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI Message-ID: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> I recently viewed an article online where IEEE just addopted 10Gb Ethernet as a standard. I belive it was 802.ae. I also know iSCSI seems to be an up and commer in the storage arena. So the comination of 10GbE and iSCSI seem be bee a pretty awesome combination. Not only iSCSE with 10GbE but also the other 802.x standards. Has anyone started playing with iSCSI? Is it in production for them? What are some of the drawbacks to iSCSI? When trying to build a *nix and *dows integrated invironment iSCSI seems to be a dream solution. -- Stephen Nordlund, MCP Systems Engineer From jim at rossberry.com Tue Mar 23 12:41:27 2004 From: jim at rossberry.com (Jim Wildman) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> Message-ID: On Tue, 23 Mar 2004, Stephen Nordlund wrote: > Has anyone started playing with iSCSI? > Is it in production for them? > What are some of the drawbacks to iSCSI? The problem with iSCSI, no matter how fast the transport, is that it is still a long serial cable. ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim@rossberry.com http://www.rossberry.com From jep200305 at columbus.rr.com Tue Mar 23 12:52:30 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> Message-ID: <20040323125230.4d290a02.jep200305@columbus.rr.com> Jim Wildman wrote: > The problem with iSCSI, no matter how fast the transport, > is that it is still a long serial cable. What are the problems with a long serial cable? That a long serial cable is still slower than a short parallel cable? That a long (serial) cable is more vulnerable than a short (parallel) cable? From jim at rossberry.com Tue Mar 23 13:38:37 2004 From: jim at rossberry.com (Jim Wildman) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: <20040323125230.4d290a02.jep200305@columbus.rr.com> References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> <20040323125230.4d290a02.jep200305@columbus.rr.com> Message-ID: On Tue, 23 Mar 2004, Jim P wrote: > What are the problems with a long serial cable? > > That a long serial cable is still slower than a short parallel cable? > That a long (serial) cable is more vulnerable than a short (parallel) cable? > Taking a parallel protocol (SCSI) and turning it sideways (iSCSI) seems to beg for increased complexity in error handling, throughput throttling, etc. There is also the issue of packet/block ordering if the host and disk are not on the same network switch, or are separated by some number of hops. I'm always leary of protocols on top of protocols on top of protocols on top of protocols. I did not mean to imply that iSCSI has no place. Just that there are a lot of issues that have to be handled by the software layer. ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim@rossberry.com http://www.rossberry.com From rfunk at funknet.net Tue Mar 23 13:44:24 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> Message-ID: <200403231344.24825.rfunk@funknet.net> Jim Wildman wrote: > The problem with iSCSI, no matter how fast the transport, is that it is > still a long serial cable. Hey, it beats a long parallel cable. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Tue Mar 23 14:00:27 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> <20040323125230.4d290a02.jep200305@columbus.rr.com> Message-ID: <20040323140027.419e7e95.jep200305@columbus.rr.com> Jim Wildman wrote: > On Tue, 23 Mar 2004, Jim P wrote: > > > What are the problems with a long serial cable? > I'm always leary of protocols on top of protocols on top of protocols on > top of protocols. Gotcha. It's not a parallel/serial issue. iSCSI over plip on a parallel printer port would be just as bad as serial iSCSI, perhaps worse. From tom at functionalmedia.com Tue Mar 23 14:00:28 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] ESR book review Message-ID: <59DB2E22-7CFC-11D8-981A-00039317745E@functionalmedia.com> The threads thread frequently included links to ESR's book "The Art of UNIX Programming" I bought a copy. It arrived last night and although I had had a long day I stayed up late and devoured the first 50 pages. This book is GREAT. 10 thumbs up !!. I really do not qualify as a programmer, web scripts are the most dangerous programs I write. But the book appeals to me as an "advanced user" . I have long struggled to explain to the non-technical why linux is a good thing. This book gives me the words to explain and understand what makes Linux so good !! A little too much focus on the ESR RMS feud. Looking back that is/was a dark episode for all of us. Perhaps not too much focus but I just hate being reminded of that dispute. Like a bad family episode that you simply do not want to be brought up again. Let bygones be bygones. The rise of linux and free/open software is a huge victory no matter how you slice it. Buy that book !! I would loan you my copy but I keep underlining and circling so mine will have to remain with me. Regards, Tom Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From pstjohn at hrblock.com Tue Mar 23 14:19:45 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] ESR book review Message-ID: Is Kernighan and Pike too programmer-oriented, or is it too old? I haven't glanced at a **book** about unix, in years; only technical docs. So I have no basis for comparison, and maybe don't know what to recommend to people. But by the standards of the day, among books about **specific** operating systems (better to say, "Development Environments"), K & Pike was just an unparalleled triumph. So maybe I answered my own question; it's about Unix as a "Programming Environment", not as a "server platform"? Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of tom hanlon > Sent: Tuesday, March 23, 2004 2:00 PM > To: colug@colug.net > Subject: [COLUG] ESR book review > > The threads thread frequently included links to ESR's book "The Art of > UNIX Programming" > > I bought a copy. It arrived last night and although I had had a long > day I stayed up late and devoured the first 50 pages. > > This book is GREAT. 10 thumbs up !!. > > I really do not qualify as a programmer, web scripts are the most > dangerous programs I write. But the book appeals to me as an "advanced > user" . > > I have long struggled to explain to the non-technical why linux is a > good thing. This book gives me the words to explain and understand what > makes Linux so good !! > > A little too much focus on the ESR RMS feud. Looking back that is/was a > dark episode for all of us. Perhaps not too much focus but I just hate > being reminded of that dispute. Like a bad family episode that you > simply do not want to be brought up again. Let bygones be bygones. The > rise of linux and free/open software is a huge victory no matter how > you slice it. > > Buy that book !! > > I would loan you my copy but I keep underlining and circling so mine > will have to remain with me. > > Regards, > > Tom > > > Tom Hanlon > Functional Media > 740-597-1472 > tom@functionalmedia.com > www.functionalmedia.com > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040323/e18b0184/attachment.htm From jim at rossberry.com Tue Mar 23 15:03:01 2004 From: jim at rossberry.com (Jim Wildman) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] iSCSI In-Reply-To: <20040323140027.419e7e95.jep200305@columbus.rr.com> References: <63708.155.188.191.5.1080056842.squirrel@mail.ilive4code.net> <20040323125230.4d290a02.jep200305@columbus.rr.com> <20040323140027.419e7e95.jep200305@columbus.rr.com> Message-ID: On Tue, 23 Mar 2004, Jim P wrote: > > I'm always leary of protocols on top of protocols on top of protocols on > > top of protocols. > > Gotcha. It's not a parallel/serial issue. > iSCSI over plip on a parallel printer port > would be just as bad as serial iSCSI, perhaps worse. So there are lots of ways to use/abuse/misuse/reuse SCSI. :-) ------------------------------------------------------------------------ Jim Wildman, CISSP, RHCE jim@rossberry.com http://www.rossberry.com From tom at functionalmedia.com Tue Mar 23 15:04:57 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] ESR book review In-Reply-To: Message-ID: <5B98936E-7D05-11D8-981A-00039317745E@functionalmedia.com> Kernighan and Pike, I have not read. I assume that for me it is too programmer oriented. Depends on who you are. The Raymond book claims to apply to advanced users and programmers. Your Milage may vary. I found the first 50 pages to be great. I also understand that the K&P book is a classic. Unix as "Programming environment" or "server platform". It appeals to me because of the second. Perhaps it is such a great server platform due to the strong programming environment under the hood. Tom On Tuesday, March 23, 2004, at 02:19 PM, St. John, Peter wrote: > Is Kernighan and Pike too programmer-oriented, or is it too old? > > I haven't glanced at a **book** about unix, in years; only technical > docs. So I have no basis for comparison, and maybe don't know what to > recommend to people. > > But by the standards of the day, among books about **specific** > operating systems (better to say, "Development Environments"), K & > Pike was just an unparalleled triumph. > > So maybe I answered my own question; it's about Unix as a "Programming > Environment", not as a "server platform"? > > Peter > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 1751 bytes Desc: not available Url : http://www.colug.net/pipermail/colug/attachments/20040323/88b8738a/attachment.bin From colug at jmglov.net Tue Mar 23 16:16:23 2004 From: colug at jmglov.net (Josh Glover) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] ESR book review In-Reply-To: <5B98936E-7D05-11D8-981A-00039317745E@functionalmedia.com> References: <5B98936E-7D05-11D8-981A-00039317745E@functionalmedia.com> Message-ID: <2919.24.123.50.150.1080076583.squirrel@mail.jmglov.net> Quoth tom hanlon: > On Tuesday, March 23, 2004, at 02:19 PM, St. John, Peter wrote: > >> Is Kernighan and Pike too programmer-oriented, or is it too old? > > Kernighan and Pike, I have not read. I assume that for me it is too > programmer oriented. Depends on who you are. The Raymond book claims to > apply to advanced users and programmers. Your Milage may vary. I found > the first 50 pages to be great. I also understand that the K&P book is > a classic. I assume Peter is talking about _The UNIX Programming Environment_, and not _The Practice of Programming_ (which any professional programmer should read, if not own). I have read both books, and while there is some overlap, TAOUP has a lot more explanation of the why involved in Unix programming conventions, and more programming best-practises stuff. And sure, it is newer. TUPE illustrates a lot of the basics, like pipelines, how to use grep, etc. TAOUP does not do much of this. Any serious Unix user, programmer or not, should read most of TUPE. Programmers should read it cover to cover. TAOUP is also highly recommended, though I would say some decent knowledge of basic Unix tools or reading TUPE should be a prerequesite. -- Josh Glover GPG keyID 0xDE8A3103 (C3E4 FA9E 1E07 BBDB 6D8B 07AB 2BF1 67A1 DE8A 3103) gpg --keyserver pgp.mit.edu --recv-keys DE8A3103 From s.molnar at sbcglobal.net Tue Mar 23 17:05:13 2004 From: s.molnar at sbcglobal.net (Stephen P. Molnar, Ph.D.) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] Apache Problem Message-ID: <20040323170513.6cf5ac9e.s.molnar@sbcglobal.net> I am running SuSE9.0/KDE3.2.1/Apache2. Suddenly, unaccountable the URL http://127.0.0.1/~chemistry/cgi-bin/webmo/login.cgi is resulting in: Server error! The server encountered an internal error and was unable to complete your request. Error message: Premature end of script headers: login.cgi If you think this is a server error, please contact the webmaster. Error 500 127.0.0.1 Tue Mar 23 17:01:30 2004 Apache/2.0.48 (Linux/SuSE) I have made no changes to the sytem. How do I restore access to the URL? Thanks in advance. -- Stephen P. Molnar, Ph.D. Life is a fuzzy set Foundation for Chemistry Stochastic and mulitvariant http://www.geocities.com/FoundationForChemistry From rfunk at funknet.net Tue Mar 23 17:18:46 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] ESR book review In-Reply-To: <2919.24.123.50.150.1080076583.squirrel@mail.jmglov.net> References: <5B98936E-7D05-11D8-981A-00039317745E@functionalmedia.com> <2919.24.123.50.150.1080076583.squirrel@mail.jmglov.net> Message-ID: <200403231718.46960.rfunk@funknet.net> Josh Glover wrote: > I assume Peter is talking about _The UNIX Programming Environment_, and > not _The Practice of Programming_ (which any professional programmer > should read, if not own). Heh, considering the context I actually made the opposite assumption. > I have read both books, and while there is some overlap, TAOUP has a lot > more explanation of the why involved in Unix programming conventions, > and more programming best-practises stuff. And sure, it is newer. TUPE > illustrates a lot of the basics, like pipelines, how to use grep, etc. > TAOUP does not do much of this. I don't think TUPE is in any way comparable to TAOUP. TUPE is a purely technical book, while TAOUP is more of a technical philosophy book. TUPE says "this is the way it is", while TAOUP says "this is why it's the way it is, this is why it's good, and this is why you should do things this way." TPOP is also more of a technical philosophy book, but has a somewhat different topical focus than TAOUP and balances a little more to the technical side. > Any serious Unix user, programmer or not, should read most of TUPE. While it's a handy book at times with a focus that isn't exactly replicated in many places, it's dated enough that I wouldn't put it on the required reading list. Enough details have changed in the past 20 years that it could be very frustrating for some people. (For example, it tells people to use # to erase the previous character.) I see TUPE as two books in one -- an introduction to using Unix circa 1983, and an introduction to basic Unix programming and software development tools. The first part is the most dated, and also the one most easily replaced with modern equivalents. In the early 90s I got the same information (but more current) from Kit Wang's "An Introduction to Berkeley Unix"; others might go for the Linux Cookbook or another current Linux book. The part on programming tools may be replaced by the decade-newer 4.4BSD programmer's reference guide. Unfortunately the published version is out of print. (It's available in online Unix archives though.) "Linux Application Development" by Michael Johnson and Erik Troan is a modern substitute, though a bit too focused on Red Hat tools at times. (Understandable given who the authors are.) Of course, as a non-programmer Tom doesn't care about any of this. :-) (Though he might be interested in "The Unix Philosophy" by Mike Gancarz.) -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From phostetl at columbus.rr.com Tue Mar 23 19:04:40 2004 From: phostetl at columbus.rr.com (Paul Hostetler) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] Apache Problem In-Reply-To: <20040323170513.6cf5ac9e.s.molnar@sbcglobal.net> References: <20040323170513.6cf5ac9e.s.molnar@sbcglobal.net> Message-ID: On Mar 23, 2004, at 5:05 PM, Stephen P. Molnar, Ph.D. wrote: [chop] > I have made no changes to the sytem. > > How do I restore access to the URL? > > Thanks in advance. What do the error logs say? Try looking in /var/log/apache2/ From tom at functionalmedia.com Tue Mar 23 19:08:15 2004 From: tom at functionalmedia.com (tom hanlon) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] Apache Problem In-Reply-To: <20040323170513.6cf5ac9e.s.molnar@sbcglobal.net> Message-ID: <58BB57E8-7D27-11D8-981A-00039317745E@functionalmedia.com> There is a common cause for this error. Let me see if I recall... If it is not this then it is some other common problem. Try the below if that fails write a very simple cgi script that works from the command line then try that in that directory as a url. This will guarantee cgi functionality. Work it out from there. First guess... I suspect that you edited the file in windows ?? Or perhaps you uploaded it in binary instead of text mode or vice versa. I believe it points to a text formatting of the script error. Undo what you did either the ftp or text edit from windows I suspect. As a sidenote... Sometimes script errors show up as a strange failure to find elf.out?? libraries. I think this is an extension of a similar problem as yours. Script fails, attempts to spew output using elf.out (??) the libraries are unavailable so instead of the script saying " I have failed" you get the script saying " I have failed to tell you that I failed" Tom On Tuesday, March 23, 2004, at 05:05 PM, Stephen P. Molnar, Ph.D. wrote: > I am running SuSE9.0/KDE3.2.1/Apache2. Suddenly, unaccountable the > > URL http://127.0.0.1/~chemistry/cgi-bin/webmo/login.cgi > > is resulting in: > > Server error! > > The server encountered an internal error and was unable to complete > your > request. > > Error message: > Premature end of script headers: login.cgi > > If you think this is a server error, please contact the webmaster. > Error 500 > 127.0.0.1 > Tue Mar 23 17:01:30 2004 > Apache/2.0.48 (Linux/SuSE) > > I have made no changes to the sytem. > > How do I restore access to the URL? > > Thanks in advance. > > -- > Stephen P. Molnar, Ph.D. Life is a fuzzy set > Foundation for Chemistry Stochastic and mulitvariant > http://www.geocities.com/FoundationForChemistry > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From s.molnar at sbcglobal.net Wed Mar 24 07:05:26 2004 From: s.molnar at sbcglobal.net (Stephen P. Molnar, Ph.D.) Date: Sat Jan 8 01:34:49 2005 Subject: [COLUG] Apache Problem In-Reply-To: <58BB57E8-7D27-11D8-981A-00039317745E@functionalmedia.com> References: <20040323170513.6cf5ac9e.s.molnar@sbcglobal.net> <58BB57E8-7D27-11D8-981A-00039317745E@functionalmedia.com> Message-ID: <6.0.0.22.2.20040324070321.028cdf38@pop.sbcglobal.yahoo.com> Thanks for the reply. As it turns out the problem was my own stupidity. WebMO has a really nice installation script which I followed blindly, selecting the wrong location for perl. All is now well. At 07:08 PM 3/23/2004, you wrote: >There is a common cause for this error. > >Let me see if I recall... If it is not this then it is some other common >problem. >Try the below if that fails write a very simple cgi script that works from >the command line then try that in that directory as a url. This will >guarantee cgi functionality. Work it out from there. > >First guess... >I suspect that you edited the file in windows ?? >Or perhaps you uploaded it in binary instead of text mode or vice versa. > >I believe it points to a text formatting of the script error. >Undo what you did either the ftp or text edit from windows I suspect. > >As a sidenote... >Sometimes script errors show up as a strange failure to find elf.out?? >libraries. I think this is an extension of a similar problem as yours. >Script fails, attempts to spew output using elf.out (??) the libraries are >unavailable so instead of the script saying " I have failed" you get the >script saying " I have failed to tell you that I failed" > > >Tom > > > > > >On Tuesday, March 23, 2004, at 05:05 PM, Stephen P. Molnar, Ph.D. wrote: > >>I am running SuSE9.0/KDE3.2.1/Apache2. Suddenly, unaccountable the >> >>URL http://127.0.0.1/~chemistry/cgi-bin/webmo/login.cgi >> >>is resulting in: >> >>Server error! >> >>The server encountered an internal error and was unable to complete your >>request. >> >>Error message: >>Premature end of script headers: login.cgi >> >>If you think this is a server error, please contact the webmaster. >>Error 500 >>127.0.0.1 >>Tue Mar 23 17:01:30 2004 >>Apache/2.0.48 (Linux/SuSE) >> >>I have made no changes to the sytem. >> >>How do I restore access to the URL? >> >>Thanks in advance. >> >>-- >>Stephen P. Molnar, Ph.D. Life is a fuzzy set >>Foundation for Chemistry Stochastic and mulitvariant >>http://www.geocities.com/FoundationForChemistry >>_______________________________________________ >>colug mailing list >>colug@colug.net >>http://www.colug.net/mailman/listinfo/colug >> >Tom Hanlon >Functional Media >740-597-1472 >tom@functionalmedia.com >www.functionalmedia.com > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug Stephen P. Molnar, Ph.D. Life is a fuzzy set Foundation for Chemistry Multivariant and stochastic http://www.geocities.com/FoundationForChemistry From ken at alpha2.com Wed Mar 24 08:57:55 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf Message-ID: <021801c411a8$01f30c20$690aa8c0@alpha2.com> Can anyone tell me the correct way to move: rmmod ide-tape modprobe ide-scsi hdparm -d0 /dev/hdt0 mt -f /dev/st0 stopoptions no-blklimits mt -f /dev/st0 stopoptions no-blklimits from /etc/rc.d/rc.local to /etc/modules.conf? Ken Bradford Alpha II Service, Inc. From ken at alpha2.com Wed Mar 24 09:48:57 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <20040324091556.18f7e768.jep200305@columbus.rr.com> Message-ID: <022501c411af$23081de0$690aa8c0@alpha2.com> > -----Original Message----- > From: Jim P [mailto:jep200305@columbus.rr.com] > Sent: Wednesday, March 24, 2004 9:16 AM > To: ken@alpha2.com > Subject: Re: [COLUG] modules.conf > Importance: High > > > On Wed, 24 Mar 2004 08:57:55 -0500 Ken Bradford wrote: > > > Can anyone tell me the correct way to move: > > > > rmmod ide-tape > > modprobe ide-scsi > > hdparm -d0 /dev/hdt0 > > mt -f /dev/st0 stopoptions no-blklimits > > mt -f /dev/st0 stopoptions no-blklimits typo, should be: stoptions > > > > from /etc/rc.d/rc.local to /etc/modules.conf? > > I don't see how the last three commands, > especially the last two commands, would go in modules.conf. I'd like to move it all to modules.conf for ease of use with mondo rescue (alternately I could investigate how to add rc.local to the mindi boot image). After reading the man page on modules.conf, I thought it would be something like: post-install ide-scsi hdparm -d0 /dev/hdt0 post-install ide-scsi mt -f /dev/st0 stoptions no-blklimits post-install ide-scsi mt -f /dev/st0 stoptions no-blklimits or post-install ide-scsi "hdparm -d0 /dev/hdt0;mt -f /dev/st0 stoptions no-blklimits;mt -f /dev/st0 stoptions no-blklimits" But my understanding is obviusly lacking because my first example generated the following errors in /var/log/messages: modprobe: modprobe: Invalid line 27 in /etc/modules.conf ^Ihdparm modprobe: modprobe: Invalid line 28 in /etc/modules.conf ^Imt modprobe: modprobe: Invalid line 29 in /etc/modules.conf ^Imt I haven't tried the second version, expecting simlar results due to my lack of understanding. > > The non-module commands would remain in /etc/rc.d/rc.local. > Alternatively, you could put the non-module commands in a > script in /etc/rc.d/init.d with soft links in the /etc/rc.d/rc?.d > directories such as below for NFS. > > [jep@tmp rc.d]$ ll */*nfs > -rwxr-xr-x 1 root root 2918 Apr 9 2002 init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc0.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc1.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc2.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc3.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc4.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc5.d/K20nfs -> ../init.d/nfs > lrwxrwxrwx 1 root root 13 Jul 26 2002 > rc6.d/K20nfs -> ../init.d/nfs > [jep@tmp rc.d]$ > > The latter is probably more hassle than you want to tangle with, > so you'll probably just leave the last few commands in rc.local. > > To maximize hassle, make a package, like an RPM. > Amen to that ;-) Ken Bradford Alpha II Service, Inc. From rfunk at funknet.net Wed Mar 24 11:05:59 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <022501c411af$23081de0$690aa8c0@alpha2.com> References: <022501c411af$23081de0$690aa8c0@alpha2.com> Message-ID: <200403241105.59610.rfunk@funknet.net> Ken Bradford wrote: > post-install ide-scsi hdparm -d0 /dev/hdt0 > post-install ide-scsi mt -f /dev/st0 stoptions no-blklimits > post-install ide-scsi mt -f /dev/st0 stoptions no-blklimits > > But my understanding is obviusly lacking because my first example > generated the following errors in /var/log/messages: > > modprobe: modprobe: Invalid line 27 in /etc/modules.conf ^Ihdparm > modprobe: modprobe: Invalid line 28 in /etc/modules.conf ^Imt > modprobe: modprobe: Invalid line 29 in /etc/modules.conf ^Imt ^I indicates a tab character. Try replacing any tabs with spaces. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From rutledge.50 at osu.edu Wed Mar 24 14:29:36 2004 From: rutledge.50 at osu.edu (LINCOLN RUTLEDGE) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] ESR book review Message-ID: <68379a67fcdb.67fcdb68379a@osu.edu> Hi, Just to throw in my two cents, I have to say that I have learned a lot from older materials this is passed over in modern books. My favorite example is "Berkeley Unix and ANSI C" from Jack Hodges. It is mainly written to be two or three college classes. It covers enough C with simple examples that I finally started "getting it". The Unix shell and basic program usage covered things that I didn't know after reading some Unix books and using Linux and BSDs for three years or so. None of my Linux books teach about job control in the shell, or how to connect to a terminal server and then log in to a shell account on a machine (although I had accidentally done it trying to get on the Web in Slackware :) ). I had used fetchmail and spools for a while, but did not know the simple usage of "mail", which would have been nice to know in my shell scripting. After reading this book I could use my 68k Mac running Linux without X, and multitask without learning "screen" (no virtual consoles in m68k) :). Very useful information that I had never come across in the man pages for the shells I've tried (although the knowledge was probably assumed in the past when you would only have one physical console to a Unix server). Anyway, I'm a wacko who likes to read info pages and source code for fun... I will have to check out these books... lincr Message: 3 Date: Tue, 23 Mar 2004 17:18:46 -0500 From: Rob Funk Subject: Re: [COLUG] ESR book review To: Central OH Linux User Group Message-ID: <200403231718.46960.rfunk@funknet.net> Content-Type: text/plain; charset="iso-8859-1" Josh Glover wrote: > I assume Peter is talking about _The UNIX Programming Environment_, and > not _The Practice of Programming_ (which any professional programmer > should read, if not own). Heh, considering the context I actually made the opposite assumption. > I have read both books, and while there is some overlap, TAOUP has a lot > more explanation of the why involved in Unix programming conventions, > and more programming best-practises stuff. And sure, it is newer. TUPE > illustrates a lot of the basics, like pipelines, how to use grep, etc. > TAOUP does not do much of this. I don't think TUPE is in any way comparable to TAOUP. TUPE is a purely technical book, while TAOUP is more of a technical philosophy book. TUPE says "this is the way it is", while TAOUP says "this is why it's the way it is, this is why it's good, and this is why you should do things this way." TPOP is also more of a technical philosophy book, but has a somewhat different topical focus than TAOUP and balances a little more to the technical side. > Any serious Unix user, programmer or not, should read most of TUPE. While it's a handy book at times with a focus that isn't exactly replicated in many places, it's dated enough that I wouldn't put it on the required reading list. Enough details have changed in the past 20 years that it could be very frustrating for some people. (For example, it tells people to use # to erase the previous character.) I see TUPE as two books in one -- an introduction to using Unix circa 1983, and an introduction to basic Unix programming and software development tools. The first part is the most dated, and also the one most easily replaced with modern equivalents. In the early 90s I got the same information (but more current) from Kit Wang's "An Introduction to Berkeley Unix"; others might go for the Linux Cookbook or another current Linux book. The part on programming tools may be replaced by the decade-newer 4.4BSD programmer's reference guide. Unfortunately the published version is out of print. (It's available in online Unix archives though.) "Linux Application Development" by Michael Johnson and Erik Troan is a modern substitute, though a bit too focused on Red Hat tools at times. (Understandable given who the authors are.) Of course, as a non-programmer Tom doesn't care about any of this. :-) (Though he might be interested in "The Unix Philosophy" by Mike Gancarz.) -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From pstjohn at hrblock.com Wed Mar 24 15:43:39 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] ESR book review Message-ID: Lincoln wrote "Anyway, I'm a wacko who likes to read info pages and source code for fun..." I like those special books that come from someone doing something cool, writing well about it; the Horses' Mouth, when the horse is articulate. Freud's Interpretation of Dreams. Caesar's Commentaries. K&R C. The source code is the final word for the here and now, of course; but a good book conveys a theme (and maybe, historical perspective) that you don't get from manuals. I'm thinking that if Ken Thompson, himself, wrote much, it would probably be poetry, not prose (:-) but I recommend the ancient original docs Dennis Ritchie scanned in to his home page, http://www.cs.bell-labs.com/who/dmr/ Peter -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040324/8f210dcd/attachment.htm From ken at alpha2.com Wed Mar 24 17:20:23 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <200403241105.59610.rfunk@funknet.net> Message-ID: <00c401c411ee$33b057e0$690aa8c0@alpha2.com> > From: Rob Funk [mailto:rfunk@funknet.net] > Ken Bradford wrote: > > But my understanding is obviusly lacking because my first example > > generated the following errors in /var/log/messages: > > > > modprobe: modprobe: Invalid line 27 in /etc/modules.conf ^Ihdparm > > modprobe: modprobe: Invalid line 28 in /etc/modules.conf ^Imt > > modprobe: modprobe: Invalid line 29 in /etc/modules.conf ^Imt > > ^I indicates a tab character. Try replacing any tabs with spaces. > > -- > ==============================| "A slice of life isn't the whole cake > Rob Funk | One tooth will never make a full grin" > http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" > I may have used tabs, as I'm now no longer logging any errors using the following lines: remove ide-tape probe ide-scsi post-install ide-scsi /sbin/hdparm -d0 /dev/hdt0 post-install ide-scsi /bin/mt -f /dev/st0 stoptions no-blklimits post-install ide-scsi /bin/mt -f /dev/st0 stoptions no-blklimits _but_ ide-scsi doesn't load and the tape drive doesn't install. Perhaps modules.conf is used too early, like _before_ ide-tape is even loaded (it's not listed there.) Perhaps something is killing ide-scsi latter on, or it's too early to load it. I'm going to check out the mondo rescue forum & archives (first) and see if I can find out how to: A) Pass along the commands, or B) Include a startup file (rc.local) in the mindi.iso, or C) find any additional advice including editing modules.conf. Ken Bradford Alpha II Service, Inc. From rfunk at funknet.net Wed Mar 24 17:43:06 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <00c401c411ee$33b057e0$690aa8c0@alpha2.com> References: <00c401c411ee$33b057e0$690aa8c0@alpha2.com> Message-ID: <200403241743.06865.rfunk@funknet.net> Ken Bradford wrote: > _but_ ide-scsi doesn't load and the tape drive doesn't install. Perhaps > modules.conf is used too early, like _before_ ide-tape is even loaded > (it's not listed there.) Perhaps something is killing ide-scsi latter > on, or it's too early to load it. I'm not too familiar with IDE tape drives, but I know that when using ide-scsi with my cd burner I need to boot the kernel with something like "hdc=scsi" to prevent the IDE drivers from grabbing it entirely. Maybe you need to do the same. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From ken at alpha2.com Thu Mar 25 08:53:26 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <200403241743.06865.rfunk@funknet.net> Message-ID: <020901c41270$8c12dca0$690aa8c0@alpha2.com> > From: Rob Funk [mailto:rfunk@funknet.net] > > I'm not too familiar with IDE tape drives, but I know that when using > ide-scsi with my cd burner I need to boot the kernel with something like > "hdc=scsi" to prevent the IDE drivers from grabbing it entirely. Maybe > you need to do the same. > > -- > ==============================| "A slice of life isn't the whole cake > Rob Funk | One tooth will never make a full grin" > http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" > Where do you do this though, because just using: rmmod ide-tape modprobe ide-scsi hdparm -d0 /dev/hdt0 mt -f /dev/st0 stoptions no-blklimits mt -f /dev/st0 stoptions no-blklimits in /etc/rc.d/rc.local makes everything right as rain. I run into problems with mondo during the compare / restore because, by default anyway, it doesn't seem to use any /etc/rc.d/ startup scripts so my tape drive is not available when booting from the mindi bootcd. Ken Bradford Alpha II Service, Inc. From s.molnar at sbcglobal.net Thu Mar 25 09:51:38 2004 From: s.molnar at sbcglobal.net (Stephen P. Molnar, Ph.D.) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Permission Problem Message-ID: <200403250951.38338.s.molnar@sbcglobal.net> I'm having a problem running an application as a user. The error messages are: ERROR ************************************************ Traceback (most recent call last): File "/usr/local/lib/CompChem/autodock/mglTools/share/lib/python2.3/site-packages/ViewerFramework/VF.py", line 563, in tryto result = apply( command, args, kw ) File "/usr/local/lib/CompChem/autodock/mglTools/share/lib/python2.3/site-packages/AutoDockTools/autodpfCommands.py", line 1185, in doit self.vf.dpo.write(outfile, genetic_algorithm_local_search_list) File "/usr/local/lib/CompChem/autodock/mglTools/share/lib/python2.3/site-packages/AutoDockTools/DockingParameters.py", line 499, in write dpf_ptr = open(filename, 'w') IOError: [Errno 13] Permission denied: '/usr/local/lib/CompChem/autodock/tutorial/ind.dpf' the user chemistry is a member of groups and the permissions of the autodock file tree are: drwxr-xr-x 14 root users 424 2004-03-20 06:37 autodock/ I would appreciate insights as to the problem and solution. Thanks in advance. -- Stephen P. Molnar, Ph.D. Life is a fuzzy set Foundation for Chemistry Multivariant and stochastic http://www.geocities.com/FoundationForChemistry From sjs at khadrin.com Thu Mar 25 09:53:43 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Permission Problem In-Reply-To: <200403250951.38338.s.molnar@sbcglobal.net> References: <200403250951.38338.s.molnar@sbcglobal.net> Message-ID: <1080226423.3830.128.camel@cobra> On Thu, 2004-03-25 at 09:51, Stephen P. Molnar, Ph.D. wrote: > "/usr/local/lib/CompChem/autodock/mglTools/share/lib/python2.3/site-packages/AutoDockTools/DockingParameters.py", > line 499, in write > dpf_ptr = open(filename, 'w') > IOError: [Errno 13] Permission denied: > '/usr/local/lib/CompChem/autodock/tutorial/ind.dpf' It looks like the application is trying to open ind.dpf for writing. Does that make sense? What are the permissions of that file? -- Stephen J. Smith | sjs@khadrin.com | http://khadrin.com/ From blitz at post891.org Thu Mar 25 10:12:45 2004 From: blitz at post891.org (Patrick Blitz) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] OpenBSD question: apache + Php recompiling? Message-ID: <1080227565.701.17.camel@amon> Hey, I'm still a steady reader, although my time available for posting to this list gets smaller and smaller. Well, I've also still a Server in Columbus, a old Sparc 5 (Pizzabox). It's running OpenBSD right now, in version 3.2. I was stupid enough to try and reinstall my own apache on this Machine. Apache worked fine, but installing php didn't work. compiling was working, but trying to restart apache with the php4 DSO didn't work. So i decided to reinstall the apache from bsd. Doesn't work. couldn't find the packages for it at all, as Apache is includede anywhere. Later, I tried to recompile Apache (1.3.27) with the config.layout set to OpenBSD. It worked good so far, but trying to install the php4-core package from OpenBSD (yeah, i know found out about it.) it gave me the following error (trying to reload apache): root@anubis :#: apachectl start Syntax error on line 236 of /var/www/conf/httpd.conf: API module structure `php4_module' in file /usr/lib/apache/modules/libphp4.so is garbled - perhaps this is not an Apache module DSO? /usr/sbin/apachectl start: httpd could not be started can anybody help me there? thanks in advance. Patrick From s.molnar at sbcglobal.net Thu Mar 25 10:58:09 2004 From: s.molnar at sbcglobal.net (Stephen P. Molnar, Ph.D.) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Permission Problem In-Reply-To: <1080226423.3830.128.camel@cobra> References: <200403250951.38338.s.molnar@sbcglobal.net> <1080226423.3830.128.camel@cobra> Message-ID: <200403251058.09830.s.molnar@sbcglobal.net> On Thursday March 25, 2004 09:53, Stephen J. Smith wrote: > On Thu, 2004-03-25 at 09:51, Stephen P. Molnar, Ph.D. wrote: > > "/usr/local/lib/CompChem/autodock/mglTools/share/lib/python2.3/site-packa > >ges/AutoDockTools/DockingParameters.py", line 499, in write > > dpf_ptr = open(filename, 'w') > > IOError: [Errno 13] Permission denied: > > '/usr/local/lib/CompChem/autodock/tutorial/ind.dpf' > > It looks like the application is trying to open ind.dpf for writing. > Does that make sense? What are the permissions of that file? This is what is happening. The application is going to write a file. This is a new file so how can it have permissions before its written? The permissions for the directory are root users and I'm running as a member of the group users. I have to admit that permissions remain a deep and dank secret to me. -- Stephen P. Molnar, Ph.D. Life is a fuzzy set Foundation for Chemistry Multivariant and stochastic http://www.geocities.com/FoundationForChemistry From ken at alpha2.com Thu Mar 25 11:05:39 2004 From: ken at alpha2.com (Ken Bradford) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] modules.conf In-Reply-To: <200403241743.06865.rfunk@funknet.net> Message-ID: <004901c41283$04489d60$690aa8c0@alpha2.com> > From: Rob Funk [mailto:rfunk@funknet.net] > I'm not too familiar with IDE tape drives, but I know that when using > ide-scsi with my cd burner I need to boot the kernel with something like > "hdc=scsi" to prevent the IDE drivers from grabbing it entirely. Maybe > you need to do the same. > > -- > ==============================| "A slice of life isn't the whole cake > Rob Funk | One tooth will never make a full grin" > http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" > I went back through my notes and checked my grub.conf and, yes, I do have /dev/ht0=ide-scsi in grub.conf. Ken Bradford Alpha II Service, Inc. From sjs at khadrin.com Thu Mar 25 11:40:22 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Permission Problem In-Reply-To: <200403251058.09830.s.molnar@sbcglobal.net> References: <200403250951.38338.s.molnar@sbcglobal.net> <1080226423.3830.128.camel@cobra> <200403251058.09830.s.molnar@sbcglobal.net> Message-ID: <1080232822.3830.185.camel@cobra> On Thu, 2004-03-25 at 10:58, Stephen P. Molnar, Ph.D. wrote: > the user chemistry is a member of groups and the permissions of the autodock > file tree are: > > drwxr-xr-x 14 root users 424 2004-03-20 06:37 autodock/ > This is what is happening. The application is going to write a file. This is > a new file so how can it have permissions before its written? The > permissions for the directory are root users and I'm running as a member of > the group users. You will need group write permission on the directory then. Only the directory you actually want to create files in will need this. You may also want to investigate the setgid bit and the sticky bit for directories, depending on the needs of the application. setgid bit - For directories on some systems, put files created in the directory into the same group as the directory, no matter what group the user who creates them is in. sticky bit - For directories on some systems, prevent users from removing or renaming a file in a directory unless they own the file or the directory; this is called the "restricted deletion flag" for the directory. > I have to admit that permissions remain a deep and dank secret to me. Try "info coreutils mode" at a bash prompt. Hope that helps. -- Stephen J. Smith | sjs@khadrin.com | http://khadrin.com/ From nliu99 at yahoo.com Thu Mar 25 16:38:03 2004 From: nliu99 at yahoo.com (Nan Liu) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: <200403251701.i2PH0WFN025945@stones.wcbe.org> Message-ID: <20040325213803.9479.qmail@web80604.mail.yahoo.com> Silly question here, but I can't seem to find a quick answer. I'm pulling my hair out trying to get sed to append \r at the end of every line in file foo: i.e. if foo contains: This is a file. change it to: This is \r a file. \r sed -e 's/$/\\r/' infile > outfile For whatever reason when I run it the \r appears at the beginning rather than the end of the line and it replaces the first 2 characters: \ris is \rfile. I tried the command on a new file I created and it's ok. I tried dos2unix on the original file and it still have the same errors. i.e. dos2unix infile | sed -e 's/$/\\r/' > outfile Any ideas? Thanks! Nan From skippy at skippy.net Thu Mar 25 16:59:40 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: <20040325213803.9479.qmail@web80604.mail.yahoo.com> References: <200403251701.i2PH0WFN025945@stones.wcbe.org> <20040325213803.9479.qmail@web80604.mail.yahoo.com> Message-ID: <39691.216.136.35.122.1080251980.squirrel@www.skippy.net> Nan Liu said: > Silly question here, but I can't seem to find a quick answer. I'm > pulling my hair out trying to get sed to append \r at the end of every > line in file foo: > > i.e. if foo contains: > > This is > a file. > > change it to: > This is \r > a file. \r > > sed -e 's/$/\\r/' infile > outfile sed -e 's/$/\\r$/' infile > outfile From jep200305 at columbus.rr.com Thu Mar 25 17:03:58 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:50 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: <20040325213803.9479.qmail@web80604.mail.yahoo.com> References: <200403251701.i2PH0WFN025945@stones.wcbe.org> <20040325213803.9479.qmail@web80604.mail.yahoo.com> Message-ID: <20040325170358.205a2fa1.jep200305@columbus.rr.com> Nan Liu wrote: > trying to get sed to append \r at the end of every > line in file foo: > I tried dos2unix Try unix2dos (without sed). From rfunk at funknet.net Thu Mar 25 17:49:52 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: <20040325213803.9479.qmail@web80604.mail.yahoo.com> References: <20040325213803.9479.qmail@web80604.mail.yahoo.com> Message-ID: <200403251749.52517.rfunk@funknet.net> Nan Liu wrote: > I tried the command on a new file I created and it's ok. I tried > dos2unix on the original file and it still have the same errors. i.e. That indicates to me that the original file has some invisible characters you're not expecting, which is confusing sed. Try running the file through "cat -tvE" to see what's going on. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jonadab at bright.net Fri Mar 26 10:40:30 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Permission Problem In-Reply-To: <200403251058.09830.s.molnar@sbcglobal.net> References: <200403250951.38338.s.molnar@sbcglobal.net> <1080226423.3830.128.camel@cobra> <200403251058.09830.s.molnar@sbcglobal.net> Message-ID: "Stephen P. Molnar, Ph.D." writes: > This is what is happening. The application is going to write a > file. This is a new file so how can it have permissions before its > written? The permissions for the directory are root users and I'm > running as a member of the group users. If it's only ever going to be this one file, you can touch the file and chmod g+w it. If the program needs permission in general to create files in the directory, then you have to give the group write permissions on the directory. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From jonadab at bright.net Fri Mar 26 10:48:36 2004 From: jonadab at bright.net (Jonadab the Unsightly One) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: <20040325213803.9479.qmail@web80604.mail.yahoo.com> References: <20040325213803.9479.qmail@web80604.mail.yahoo.com> Message-ID: Nan Liu writes: > Silly question here, but I can't seem to find a quick answer. I'm > pulling my hair out trying to get sed to append \r at the end of every > line in file foo: > > i.e. if foo contains: > > This is > a file. > > change it to: > This is \r > a file. \r You do realise that you appended " \r" not just "\r", right? > Any ideas? Thanks! I would just do this: cat infile | perl -e 'while(){ chomp; print "$_ \r\n" }' > outfile There are ways to golf that down to a much shorter line, but this way is easy to read and see what it's doing, assuming you know what $_ is. -- $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ From pstjohn at hrblock.com Fri Mar 26 11:27:39 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Trivial sed Question Message-ID: I think Nan is fixed OK, but we're unclear about what happened and why. 1. Nan is putting a backslash, followed by a lower case r, at the ends of lines. This is apparently deliberate; it's a beautiful coincidence that '\r' is the carriage-return byte. 2. The input had carriage-return, line-feed pairs at the end of each line (DOS style). The carriage return was getting read (moving the cursor to the beginning of the line) and then the linefeed replaced (overwriting the first two bytes of the line, with the backslash and the r). 3. I'm unclear if a) the Dos2unix tool just didn't work, or b) the syntax in sed was wrong; you substitute at the end of a line Nan's way, in vi, but the stream editor may necessarily have a differernt convention for that, and I haven't used sed in 20 years. So I'd appreciate a clarification from someone on that point. Thanks, Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Jonadab the Unsightly One > Sent: Friday, March 26, 2004 10:49 AM > To: Central OH Linux User Group > Subject: Re: [COLUG] Trivial sed Question > > Nan Liu writes: > > > Silly question here, but I can't seem to find a quick answer. I'm > > pulling my hair out trying to get sed to append \r at the end of every > > line in file foo: > > > > i.e. if foo contains: > > > > This is > > a file. > > > > change it to: > > This is \r > > a file. \r > > You do realise that you appended " \r" not just "\r", right? > > > Any ideas? Thanks! > > I would just do this: > cat infile | perl -e 'while(){ chomp; print "$_ \r\n" }' > outfile > > There are ways to golf that down to a much shorter line, but this way > is easy to read and see what it's doing, assuming you know what $_ is. > > -- > $;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}} > split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/ > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040326/25cfe2a3/attachment.htm From sjs at khadrin.com Fri Mar 26 11:52:49 2004 From: sjs at khadrin.com (Stephen J. Smith) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Trivial sed Question In-Reply-To: References: Message-ID: <1080319969.20953.4.camel@cobra.khadrin.com> On Fri, 2004-03-26 at 11:27, St. John, Peter wrote: > 3. I'm unclear if a) the Dos2unix tool just didn't work, or b) the It was that the dos2unix tool does inplace fixup when given a filename, rather than writing the result to stdout. Therefore the command that was tried, "dos2unix infile | sed -e 's/$/\\r/' > outfile", did not work. Without a filename parameter, dos2unix reads from stdin and writes to stdout. -- Stephen J. Smith | sjs@khadrin.com | http://khadrin.com/ From nliu99 at yahoo.com Fri Mar 26 14:55:32 2004 From: nliu99 at yahoo.com (Nan Liu) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect In-Reply-To: <200403261700.i2QH0SFN002097@stones.wcbe.org> Message-ID: <20040326195532.28076.qmail@web80604.mail.yahoo.com> Thanks for all the replies regarding sed, it was the pesky windows ^M that was causing the problem. dos2unix infile tempfile and then running sed on tempfile resolved it. Anyhow the reason I was messing with the \r is because I was using expect to send emails of the server status log to several people offsite. I didn't enable sendmail because it's a mess to configure/secure and there is no need for recieving emails on the server. It would be nice to send the files as attachments, but I don't have a MTA configured and running on the server. Any suggestions/ideas? Am I completely ignoring the obvious solution here? Thanks again! Nan From blitz at post891.org Fri Mar 26 15:07:39 2004 From: blitz at post891.org (Patrick Blitz) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect In-Reply-To: <20040326195532.28076.qmail@web80604.mail.yahoo.com> References: <20040326195532.28076.qmail@web80604.mail.yahoo.com> Message-ID: <1080331659.708.29.camel@amon> On Fri, 2004-03-26 at 20:55, Nan Liu wrote: > Thanks for all the replies regarding sed, it was the pesky windows ^M > that was causing the problem. dos2unix infile tempfile and then running > sed on tempfile resolved it. > > Anyhow the reason I was messing with the \r is because I was using > expect to send emails of the server status log to several people > offsite. I didn't enable sendmail because it's a mess to > configure/secure and there is no need for recieving emails on the > server. > > It would be nice to send the files as attachments, but I don't have a > MTA configured and running on the server. Any suggestions/ideas? Am I > completely ignoring the obvious solution here? > > Thanks again! Simplesmtp is a complete MTA, just without receiving. You can send mails, and even define some aliases. From pstjohn at hrblock.com Fri Mar 26 15:23:04 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect Message-ID: You guys don't use sendmail on servers for stuff like error notifications? I've use sendmail on NT for that, when I surely didn't want to install Outlook, much less Exchange. It's the same C program except for the socket thing. Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Patrick Blitz > Sent: Friday, March 26, 2004 3:08 PM > To: Central OH Linux User Group > Subject: Re: [COLUG] SMTP hack with expect > > On Fri, 2004-03-26 at 20:55, Nan Liu wrote: > > Thanks for all the replies regarding sed, it was the pesky windows ^M > > that was causing the problem. dos2unix infile tempfile and then running > > sed on tempfile resolved it. > > > > Anyhow the reason I was messing with the \r is because I was using > > expect to send emails of the server status log to several people > > offsite. I didn't enable sendmail because it's a mess to > > configure/secure and there is no need for recieving emails on the > > server. > > > > It would be nice to send the files as attachments, but I don't have a > > MTA configured and running on the server. Any suggestions/ideas? Am I > > completely ignoring the obvious solution here? > > > > Thanks again! > > Simplesmtp is a complete MTA, just without receiving. > > You can send mails, and even define some aliases. > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040326/b1b43c91/attachment.htm From blitz at post891.org Fri Mar 26 15:34:39 2004 From: blitz at post891.org (Patrick Blitz) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect In-Reply-To: References: k.net> Message-ID: <1080333279.704.32.camel@amon> On Fri, 2004-03-26 at 21:23, St. John, Peter wrote: > You guys don't use sendmail on servers for stuff like error > notifications? I've use sendmail on NT for that, when I surely didn't > want to install Outlook, much less Exchange. It's the same C program > except for the socket thing. > > Peter Sendmail is only useful when you want to receive mail. For sending out mail, it would be a lot easier to use something smaller than that. I use ssmtp or postfix for such purposes and sendmail for serious mail servers. i haven't used any sendmail-like programms on NT.. how does that work? Patrick From pstjohn at hrblock.com Fri Mar 26 16:12:09 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect Message-ID: I think I've found the confusion. The Sendmail I'm remembering is a command-line utility that composes stuff in SMTP format, opens the port, sends your message. There is no daemon, nothing running to receive mail. Any security issue would be the wrong thing running that program, e.g. for a virus to spread itself outwards. I went to the MAN pages on UGU (which are hard to navigate, they keep adding layers of cruft, presumably trying to stay afloat financially) and at : http://www.softlab.ntua.gr/cgi-bin/man-cgi?00+00 I foiund both what I remembered, and the probable reason you guys interpret sendmail differently: Sendmail sends a message to one or more recipients, rout ing the message over whatever networks are necessary but then among the options: -bd Run as a daemon. This requires Berkeley IPC. Sendmail will fork and run in background listening on socket 25 for incoming SMTP connections. This is normally run from /etc/rc. That is new to me. I've never run sendmail as a daemon, and didn't know it was possible. I would say, don't do that if you just want the command line facility (:-) Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of Patrick Blitz > Sent: Friday, March 26, 2004 3:35 PM > To: Central OH Linux User Group > Subject: RE: [COLUG] SMTP hack with expect > > On Fri, 2004-03-26 at 21:23, St. John, Peter wrote: > > You guys don't use sendmail on servers for stuff like error > > notifications? I've use sendmail on NT for that, when I surely didn't > > want to install Outlook, much less Exchange. It's the same C program > > except for the socket thing. > > > > Peter > > Sendmail is only useful when you want to receive mail. For sending out > mail, it would be a lot easier to use something smaller than that. > I use ssmtp or postfix for such purposes and sendmail for serious mail > servers. > i haven't used any sendmail-like programms on NT.. how does that work? > > Patrick > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040326/8ea3f4d1/attachment.htm From rfunk at funknet.net Fri Mar 26 19:53:31 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] SMTP hack with expect In-Reply-To: References: Message-ID: <200403261953.31442.rfunk@funknet.net> St. John, Peter wrote: > -bd Run as a daemon. This requires Berkeley IPC. Sendmail will fork and > run in background listening on socket 25 for incoming SMTP connections. > This is normally run from /etc/rc. That's the usual way of running sendmail. > I would say, don't do that if you just want the command line facility It still requires painful configuration. Exim is easier to set up, and simpler programs such as ssmtp are even better for send-only configuration. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Fri Mar 26 23:16:03 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] UNIX Network Programming, Volume 2 Message-ID: <20040326231603.0f3a643d.jep200305@columbus.rr.com> Blatant commercial content UNIX Network Programming Volume 2: Interprocess Communications by W. Richard Stevens Second Edition International Edition paperback for sale $30. New, unused. (For hardcover, list is $66,Amazon wants $60.06 and bookpool wants $58.95) Jim Prior From garyiddings at yahoo.com Sun Mar 28 20:32:25 2004 From: garyiddings at yahoo.com (Gary Iddings) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Anti-Productivity Pod Furniture Message-ID: <20040329013225.85875.qmail@web41905.mail.yahoo.com> The insurance company at 3590 Twin Creek Drive (first traffic light on Wilson Rd North of I-70) is getting all new furniture and throwing out the perfectly good old stuff. It's Kimball brand anti-productivity pod stuff. Much counter top with a nice strip at the back for running cables. Under counter cabinet (big bottom drawer for hanging files, two smaller drawers, superb ball bearing draw slides). Lateral(?) files (two big wide (36"?) drawers, each for sideways hanging files. Overhead shelf for stuff with light underneath. They had too much for the dumpster, so much is sitting on the parking lot waiting for a new dumpster. They have much more of the building yet to upgrade. There are plastic tabs/latches at the front of the drawer slides that release the drawer from the slides. All this stuff is heavy particle board construction. You will need to remove the drawers to budge the cabinets. __________________________________ Do you Yahoo!? Yahoo! Finance Tax Center - File online. File on time. http://taxes.yahoo.com/filing.html From nliu99 at yahoo.com Mon Mar 29 12:31:17 2004 From: nliu99 at yahoo.com (Nan Liu) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Opensource cPanel? In-Reply-To: <200403291700.i2TH0RFN031359@stones.wcbe.org> Message-ID: <20040329173117.47970.qmail@web80605.mail.yahoo.com> Hi! I'm looking at doing webhosting for a few friends and splitting the cost of owning a webserver. I'm looking for an opensource alternative to cPanel, where users can control their own domain/mail account/mysql database. Webmin looks promising, but it's for managing the entire server, I'd like individual users to be able to create mail accounts for their virtual domain and not see other people's info. Anyone have any experience using webmin or similar software? Thanks! Nan From pat at linuxcolumbus.com Mon Mar 29 13:05:18 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Opensource cPanel? In-Reply-To: <20040329173117.47970.qmail@web80605.mail.yahoo.com> References: <200403291700.i2TH0RFN031359@stones.wcbe.org> <20040329173117.47970.qmail@web80605.mail.yahoo.com> Message-ID: <20040329180518.GF12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 09:31:17AM -0800, Nan Liu wrote: > Webmin looks promising, but it's for managing the entire server, I'd > like individual users to be able to create mail accounts for their > virtual domain and not see other people's info. Anyone have any > experience using webmin or similar software? > I use qmail/vmailmgr/omail-admin. Works great. http://www.lifewithqmail.org http://untroubled.org http://omain.omnis.ch Pat From jep200305 at columbus.rr.com Mon Mar 29 13:18:20 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <20040329173117.47970.qmail@web80605.mail.yahoo.com> References: <200403291700.i2TH0RFN031359@stones.wcbe.org> <20040329173117.47970.qmail@web80605.mail.yahoo.com> Message-ID: <20040329131820.490223f9.jep200305@columbus.rr.com> Nan Liu wrote: > ... webhosting ... > I'm looking for an opensource alternative to cPanel, > where users can control their own domain/mail account/mysql > database. pat@linuxcolumbus.com replied: > I use qmail... > http://www.lifewithqmail.org Is qmail open source? This seems to be some controversy about this. From pat at linuxcolumbus.com Mon Mar 29 13:33:24 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:51 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <20040329131820.490223f9.jep200305@columbus.rr.com> References: <200403291700.i2TH0RFN031359@stones.wcbe.org> <20040329173117.47970.qmail@web80605.mail.yahoo.com> <20040329131820.490223f9.jep200305@columbus.rr.com> Message-ID: <20040329183324.GG12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 01:18:20PM -0500, Jim P wrote: > > Is qmail open source? This seems to be some controversy about this. > I will answer questions about how to install and run qmail. I will not answer questions that are better explained at http://cr.yp.to/qmail.html. Pat From pstjohn at hrblock.com Mon Mar 29 14:55:11 2004 From: pstjohn at hrblock.com (St. John, Peter) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? Message-ID: I've just gone through that site with great interest (Dr. Bernstein, the author of qmail, is a number theorist) but I found no description of the source tree; it looks like you have to download the package (tar) and see what you get. The rules about distribution are custom, at http://cr.yp.to/qmail/dist.html. So I'd say the "open source" is a good question. Peter > -----Original Message----- > From: colug-bounces@colug.net [SMTP:colug-bounces@colug.net] On Behalf Of pat@linuxcolumbus.com > Sent: Monday, March 29, 2004 1:33 PM > To: Central OH Linux User Group > Subject: Re: [COLUG] Opensource cPanel? Is qmail opensource? > > On Mon, Mar 29, 2004 at 01:18:20PM -0500, Jim P wrote: > > > > Is qmail open source? This seems to be some controversy about this. > > > > I will answer questions about how to install and run qmail. I will not > answer questions that are better explained at http://cr.yp.to/qmail.html. > > Pat > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.colug.net/pipermail/colug/attachments/20040329/8dcfcab7/attachment.htm From pat at linuxcolumbus.com Mon Mar 29 15:08:22 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: References: Message-ID: <20040329200822.GH12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 01:55:11PM -0600, St. John, Peter wrote: > I've just gone through that site with great interest (Dr. Bernstein, the author of qmail, is a number theorist) but I found no description of the source tree; it looks like you have to download the package (tar) and see what you get. The rules about distribution are custom, at http://cr.yp.to/qmail/dist.html. > So I'd say the "open source" is a good question. Inside the tar file you will find the source could that you must compile. If you want a distribution with suggested patches go to http://www.qmail.org/netqmail/ . With vmailmgr and omail-admin you get an extremely nice virtual email hosting system. Pat From skippy at skippy.net Mon Mar 29 15:21:40 2004 From: skippy at skippy.net (Scott Merrill) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: References: Message-ID: <40688554.50603@skippy.net> St. John, Peter wrote: > I've just gone through that site with great interest (Dr. Bernstein, the author of qmail, is a number theorist) but I found no description of the source tree; it looks like you have to download the package (tar) and see what you get. The rules about distribution are custom, at http://cr.yp.to/qmail/dist.html. > So I'd say the "open source" is a good question. http://www.gnu.org/licenses/license-list.html "Daniel Bernstein's licenses These licenses are not free software licenses because they do not permit publication of modified versions." Of course the FSF isn't the only group with an opinion on "free" and/or "open" software. From pat at linuxcolumbus.com Mon Mar 29 15:18:35 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <40688554.50603@skippy.net> References: <40688554.50603@skippy.net> Message-ID: <20040329201835.GI12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 03:21:40PM -0500, Scott Merrill wrote: > Of course the FSF isn't the only group with an opinion on "free" and/or > "open" software. Amen to that. Pat From rfunk at funknet.net Mon Mar 29 15:25:10 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: References: Message-ID: <200403291525.10083.rfunk@funknet.net> St. John, Peter wrote: > The rules about distribution are custom, at > http://cr.yp.to/qmail/dist.html. So I'd say the "open source" is a good > question. Opensource.org does not list the qmail license in its list of approved open-source licenses. http://www.opensource.org/licenses/ The FSF says: http://www.gnu.org/licenses/license-list.html#NonFreeSoftwareLicense | Daniel Bernstein's licenses | These licenses are not free software licenses because they do not permit | publication of modified versions. OpenBSD avoids Bernstein's software for the same reason and more. http://archives.neohapsis.com/archives/openbsd/2001-08/2544.html Also Debian: http://lists.debian.org/debian-devel/1997/debian-devel-199702/msg00412.html http://lists.debian.org/debian-vote/2004/debian-vote-200401/msg01066.html | qmail-src djb license. no distribution of modified versions. | no binary distribution. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From pat at linuxcolumbus.com Mon Mar 29 15:55:38 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <200403291525.10083.rfunk@funknet.net> References: <200403291525.10083.rfunk@funknet.net> Message-ID: <20040329205538.GK12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 03:25:10PM -0500, Rob Funk wrote: > Also Debian: > http://lists.debian.org/debian-devel/1997/debian-devel-199702/msg00412.html > http://lists.debian.org/debian-vote/2004/debian-vote-200401/msg01066.html > | qmail-src djb license. no distribution of modified versions. > | no binary distribution. > Information about the qmail deb packages are available here. http://packages.debian.org/stable/mail/qmail-src from which I quote: qmail is a secure Secure, reliable, efficient, simple mail transport system. Dan Bernstein (qmail's author) only gives permission for qmail to be distributed in source form, or binary for by approval. This package has been put together to allow people to easily build a qmail binary package for themselves, from source. Install and enjoy! Pat This message brought to you by qmail. :) From rfunk at funknet.net Mon Mar 29 16:13:46 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <20040329205538.GK12292@linuxcolumbus.com> References: <200403291525.10083.rfunk@funknet.net> <20040329205538.GK12292@linuxcolumbus.com> Message-ID: <200403291613.46424.rfunk@funknet.net> pat@linuxcolumbus.com wrote: > http://packages.debian.org/stable/mail/qmail-src from which I quote: > > qmail is a secure Secure, reliable, efficient, simple mail transport > system. > > Dan Bernstein (qmail's author) only gives permission for qmail to be > distributed in source form, or binary for by approval. This package has > been put together to allow people to easily build a qmail binary > package for themselves, from source. You forgot to quote a couple of pieces: | Package: qmail-src (1.03-24) [non-free] ^^^^^^^^ | Source only package for building qmail binary package ... | This package builds a binary .deb that is FHS compliant and conforms to | the Debian standards guidelines. The resulting binary packages are not | suitable for re-distribution. Non-free is not officially part of Debian. It's made up of packages that are free enough to be distributed by the Debian mirrors, but do not meet the Debian Free Software Guidelines. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From pat at linuxcolumbus.com Mon Mar 29 16:38:19 2004 From: pat at linuxcolumbus.com (pat@linuxcolumbus.com) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <200403291613.46424.rfunk@funknet.net> References: <200403291525.10083.rfunk@funknet.net> <20040329205538.GK12292@linuxcolumbus.com> <200403291613.46424.rfunk@funknet.net> Message-ID: <20040329213819.GL12292@linuxcolumbus.com> On Mon, Mar 29, 2004 at 04:13:46PM -0500, Rob Funk wrote: > Non-free is not officially part of Debian. It's made up of packages that > are free enough to be distributed by the Debian mirrors, but do not meet > the Debian Free Software Guidelines. > Non-free...free enough...Free Software. Which is it? They seem be having some issues with exactly what free is. That is why I recommand you make the decision for your self and not let all those distracters get in the way of your own free will. BTW, My favorite way to install qmail is by hand. There is just someting full filling knowing that I won't have to be watching bugtraq like a hawk. Anyway, do you have a problem with installing are running qmail that I can help you with? It is an excellent program. I use it locally with just qmail-send running and my smtproutes file pointing to my actual mail server. And all it took was "emerge qmail" and "echo ":mysmtpserver" >> /var/qmail/control/smtproutes"! It doesn't get any easier than that. Pat From rfunk at funknet.net Mon Mar 29 16:56:17 2004 From: rfunk at funknet.net (Rob Funk) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel? Is qmail opensource? In-Reply-To: <20040329213819.GL12292@linuxcolumbus.com> References: <200403291613.46424.rfunk@funknet.net> <20040329213819.GL12292@linuxcolumbus.com> Message-ID: <200403291656.17268.rfunk@funknet.net> pat@linuxcolumbus.com wrote: > On Mon, Mar 29, 2004 at 04:13:46PM -0500, Rob Funk wrote: > > Non-free is not officially part of Debian. It's made up of packages > > that are free enough to be distributed by the Debian mirrors, but do > > not meet the Debian Free Software Guidelines. > > Non-free...free enough...Free Software. Which is it? They seem be having > some issues with exactly what free is. Not at all. As far as Debian is concerned, "what free is" is defined by the Debian Free Software Guidelines (DFSG). http://www.debian.org/social_contract.html#guidelines > That is why I recommand you make > the decision for your self and not let all those distracters get in the > way of your own free will. What decision? If you're talking about the decision of what software to run, then I agree. But if you're talking about the "decision" of whether qmail is free software, then I vehemently disagree. "Open source" and "free software" have specific definitions. You don't have to limit yourself to software that meets those definitions, but please don't confuse the issue further by denying that those definitions exist or implying that these definitions are somehow invalid. http://www.gnu.org/philosophy/free-sw.html http://www.opensource.org/docs/definition.php (derived from the DFSG) > Anyway, do you have a problem with installing are running qmail that I > can help you with? Nope. -- ==============================| "A slice of life isn't the whole cake Rob Funk | One tooth will never make a full grin" http://www.funknet.net/rfunk | -- Chris Mars, "Stuck in Rewind" From jep200305 at columbus.rr.com Mon Mar 29 17:01:10 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource Control Panl for Qmail (getting back toward Nan Liu's issue) In-Reply-To: <20040329213819.GL12292@linuxcolumbus.com> References: <200403291525.10083.rfunk@funknet.net> <20040329205538.GK12292@linuxcolumbus.com> <200403291613.46424.rfunk@funknet.net> <20040329213819.GL12292@linuxcolumbus.com> Message-ID: <20040329170110.68f91a2a.jep200305@columbus.rr.com> Pat, What opensource alternative to http://cpanel.net/ control panel do you recommend so that users who aren't trusted with root privileges can (perhaps indirectly) administer qmail? From jep200305 at columbus.rr.com Mon Mar 29 17:09:52 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Opensource cPanel: http://webcp.can-host.com/ In-Reply-To: <20040329173117.47970.qmail@web80605.mail.yahoo.com> References: <200403291700.i2TH0RFN031359@stones.wcbe.org> <20040329173117.47970.qmail@web80605.mail.yahoo.com> Message-ID: <20040329170952.0e8bd6a7.jep200305@columbus.rr.com> Nan Liu wrote: > I'm looking for an opensource alternative to cPanel, > where users can control their own domain/mail account/mysql database. > Anyone have any experience using webmin or similar software? I have no experience with http://webcp.can-host.com/. From anderson at wks.uts.ohio-state.edu Mon Mar 29 20:21:12 2004 From: anderson at wks.uts.ohio-state.edu (Bill Anderson) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Windows? :/ Message-ID: I'm managing a network of 40-50 Windows machines (mostly XP, some 2K) as clients with FreeBSD samba servers/firewalls. I'm wondering if anyone has seen/used any free/open software which makes windows administration easier. Some things I'm looking for are: A tool which allows 1 image of an XP box to be pushed out to all the clients on bootup (I guess Norton Ghost Enterprise Edition does this?) (and any info from anyone who as done this) Something to make patch management easier Something to make installation of software on all machines easier Something to make configuration changes on all machines easier I'm dreading having to make complicated batch files to do things, or worse yet individually logging in to each machine and wearing out my mouse (and arm, for that matter) to accomplish things. Yes. I would love to get all of these machines running Linux/BSD, but I'm afraid it won't happen for some time, due to proprietary mission-critical software. Perhaps I'll be able to move them to WINE at some point, but it will be a while. Bill From jep200305 at columbus.rr.com Mon Mar 29 20:28:08 2004 From: jep200305 at columbus.rr.com (Jim P) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Windows? :/ In-Reply-To: References: Message-ID: <20040329202808.3958b3d0.jep200305@columbus.rr.com> Bill Anderson wrote: > A tool which allows 1 image of an XP box to be pushed out to all the > [Windows] clients on bootup Do you mean that you want to essentially reinstall Windows on each Windows box each time it boots? From debateable at ameritech.net Mon Mar 29 21:17:37 2004 From: debateable at ameritech.net (Jason Carr) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Windows? :/ In-Reply-To: References: Message-ID: <4068D8C1.1030306@ameritech.net> Do you mean something like the windows boxes at OSU that refresh the system and all every time they boot? That way if someone wrecks the box it just needs rebooted? I know they work something like that...maybe they'd know. Bill Anderson wrote: >I'm managing a network of 40-50 Windows machines (mostly XP, some 2K) as >clients with FreeBSD samba servers/firewalls. I'm wondering if anyone has >seen/used any free/open software which makes windows administration >easier. Some things I'm looking for are: > >A tool which allows 1 image of an XP box to be pushed out to all the >clients on bootup (I guess Norton Ghost Enterprise Edition does this?) >(and any info from anyone who as done this) > >Something to make patch management easier > >Something to make installation of software on all machines easier > >Something to make configuration changes on all machines easier > > >I'm dreading having to make complicated batch files to do things, or worse >yet individually logging in to each machine and wearing out my mouse (and >arm, for that matter) to accomplish things. > > >Yes. I would love to get all of these machines running Linux/BSD, but I'm >afraid it won't happen for some time, due to proprietary mission-critical >software. Perhaps I'll be able to move them to WINE at some point, but it >will be a while. > > >Bill > >_______________________________________________ >colug mailing list >colug@colug.net >http://www.colug.net/mailman/listinfo/colug > > > From tomjhanlon at earthlink.net Mon Mar 29 21:33:40 2004 From: tomjhanlon at earthlink.net (Tom Hanlon) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Windows? :/ In-Reply-To: Message-ID: Over the past two months sysadmin magazine has included the MS toolkit for Unix. Windows services for UNIX.. May or may not have anything to do with your task but an interesting product on many levels. Tom On Monday, March 29, 2004, at 08:21 PM, Bill Anderson wrote: > I'm managing a network of 40-50 Windows machines (mostly XP, some 2K) > as > clients with FreeBSD samba servers/firewalls. I'm wondering if anyone > has > seen/used any free/open software which makes windows administration > easier. Some things I'm looking for are: > > A tool which allows 1 image of an XP box to be pushed out to all the > clients on bootup (I guess Norton Ghost Enterprise Edition does this?) > (and any info from anyone who as done this) > > Something to make patch management easier > > Something to make installation of software on all machines easier > > Something to make configuration changes on all machines easier > > > I'm dreading having to make complicated batch files to do things, or > worse > yet individually logging in to each machine and wearing out my mouse > (and > arm, for that matter) to accomplish things. > > > Yes. I would love to get all of these machines running Linux/BSD, but > I'm > afraid it won't happen for some time, due to proprietary > mission-critical > software. Perhaps I'll be able to move them to WINE at some point, > but it > will be a while. > > > Bill > > _______________________________________________ > colug mailing list > colug@colug.net > http://www.colug.net/mailman/listinfo/colug > > Tom Hanlon Functional Media 740-597-1472 tom@functionalmedia.com www.functionalmedia.com From blata at extent0006.entomology.ohio-state.edu Mon Mar 29 21:38:50 2004 From: blata at extent0006.entomology.ohio-state.edu (Wade Pinkston) Date: Sat Jan 8 01:34:52 2005 Subject: [COLUG] Windows? :/ In-Reply-To: References: Message-ID: <4068DDBA.1030708@bugs.osu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 every thing you are talking about can be done with ActiveDirectory (except for "1 image of an XP box to be pushed out to all"). Active directory is part of 2k pro. you put all the winblows machines in the domain then using the 2k server you can push patches, etc etc etc. You can even turn off access to process and apps like IE, outlook. and any other programs you want. Not to mention it make s you feel like an all powerful overlord knowing you can control every thing those 40-50 other machines can do. You wrote this to me on 03/29/2004 08:21 PM: | I'm managing a network of 40-50 Windows machines (mostly XP, some | 2K) as clients with FreeBSD samba servers/firewalls. I'm wondering | if anyone has seen/used any free/open software which makes windows | administration easier. Some things I'm looking for are: | | A tool which allows 1 image of an XP box to be pushed out to all | the clients on bootup (I guess Norton Ghost Enterprise Edition does | this?) (and any info from anyone who as done this) | | Something to make patch management easier | | Something to make installation of software on all machines easier | | Something to make configuration changes on all machines easier | | | I'm dreading having to make complicated batch files to do things, | or worse yet individually logging in to each machine and wearing | out my mouse (and arm, for that matter) to accomplish things. | | | Yes. I would love to get all of these machines running Linux/BSD, | but I'm afraid it won't happen for some time, due to proprietary | mission-critical software. Perhaps I'll be able to move them to | WINE at some point, but it will be a while. | | | Bill | | _______________________________________________ colug mailing list | colug@colug.net http://www.colug.net/mailman/listinfo/colug - -- Wade Pinkston The Ohio State University Extension Entomology 1991 Kenny Rd Columbus OH 43210 phone: (614) 292-5274 Ipsa scientia potestas est Windows,a 32 bit graphical interface for a 16 bit patch to an 8 bit operating system internally coded for a 4 bit processor written by a 2 bit company that can't stand 1 bit of competition GnuPG Key ID 0x216FDD35 gpg --keyserver pgp.mit.edu --recv-keys 216FDD35 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFAaN25v+6+qSFv3TURAo5CAKC4j4Q1wsebjIcVSABj91y0ivi/LACgmkqe jqljdrtdS866ZbnOmP4Tc5E= =3vZH -----END PGP SIGNATURE----- From peter.king at utoronto.ca Tue Mar 30 00:24:54 2004 From: peter.king at utoronto.ca (Peter King) Date: Sat Jan 8 01:34:53 2005 Subject: [COLUG] Mozilla Fonts... Message-ID: <20040330052454.GA23342@socrates> I've done it again. I upgraded to XFree86 4.3 (Debian Testing), and, not surprisingly, a thing or two broke along the way. Most of them weren't too hard to fix. But this one is giving me fits, because I remember *that* I solved it the last time I upgraded, but for the life of me I don't remember just *how*... Here's the deal: in Mozilla, the only fonts available for use are two, "Serif" and "Sans-Serif". I can't figure out how to make the other installed fonts available to it. I have a hazy memory that the last time this happened it had to do with a conflict between OpenOffice and Mozilla, and that by simply deleting some OO fontfile or other the whole problem went away. But I can't remember, and I can't find it by ordinary Googling the web. Aaarggghhh. Anyone know the answer? I've added font paths to XF86Config-4, run xprint and xfs, and fiddled with every preference file I found. No joy. -- Peter King peter.king@utoronto.ca Department of Philosophy 215 Huron Street The University of Toronto (416)-978-3788 ofc Toronto, ON M5S 1A1 CANADA http://individual.utoronto.ca/pking/ ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://www.colug.net/pipermail/colug/attachments/20040330/b9d6a3d7/attachment.bin From cacepi at adelphia.net Tue Mar 30 02:01:42 2004 From: cacepi at adelphia.net (Jason Martens) Date: Sat Jan 8 01:34:53 2005 Subject: [COLUG] Mozilla Fonts... In-Reply-To: <20040330052454.GA23342@socrates> References: <20040330052454.GA23342@socrates> Message-ID: <40691B56.7040109@adelphia.net> Peter King wrote: > I've done it again. > > Here's the deal: in Mozilla, the only fonts available for use are > two, "Serif" and "Sans-Serif". I can't figure out how to make the > other installed fonts available to it. > > Anyone know the answer? I've added font paths to XF86Config-4, run > xprint and xfs, and fiddled with every preference file I found. No > joy. XFree 4.3.0 also uses fontconfig, so you'll need to make sure the font paths in /etc/X11/XF86Config is available to fontconfig. I don't use Debian anymore, but it seems most other distros place the fontconfig stuff in /etc/fonts. Make sure you edit fonts.local to keep changes persistent. Hope this helps, Jason Martens cacepi@adelphia.net From beano21 at direcway.com Tue Mar 30 06:21:14 2004 From: beano21 at direcway.com (Frank Rieder) Date: Sat Jan 8 01:34:53 2005 Subject: [COLUG] Quirk PPC In-Reply-To: <40691B56.7040109@adelphia.net> Message-ID: Anyone have experience with booting oldworld macs into Linux without the mac os? Using Quirk or something like it? At this point booting from floppy all of the time would be ok. I have an old 8500/120 with a 180MP card, and would love to get this running Linux. For now I am reinstalling the Mac OS (because I enabled the first partition which wiped the partition info) and will be installing Linux from there. Anyone played with these before? From dmaxwell at columbus.rr.com Tue Mar 30 07:05:13 2004 From: dmaxwell at columbus.rr.com (Dave Maxwell) Date: Sat Jan 8 01:34:53 2005 Subject: [COLUG] Quirk PPC In-Reply-To: References: Message-ID: <200403300705.13434.dmaxwell@columbus.rr.com>