[COLUG] FW: Debian generated SSH-Keys working exploit

[Duane]

Warner Moore wrote:
> 	Here's a reality check.  What a stupid mistake on Debian's part. 
> 
>> -----Original Message-----
>> From: mm at deadbeef.de [mailto:mm at deadbeef.de] 
>> Sent: Thursday, May 15, 2008 1:54 AM
>> To: bugtraq at securityfocus.com
>> Subject: Debian generated SSH-Keys working exploit
>>
>> Hi Securityfocus,
>>
>>
>> the debian openssl issue leads that there are only 65.536 
>> possible ssh keys generated, cause the only entropy is the 
>> pid of the process generating the key.

That's not entirely accurate, the debian/ubuntu have been distributing a
blacklist of 2k RSA keys and 1k DSA keys, the files contain 98,314
lines, also someone generated all keys for 1024, 2048 and 4096 bit.

http://metasploit.com/users/hdm/tools/debian-openssl/

Using 31 cores on a Xeon processor, it took 2 hrs for the 1024 + 2048
bit keys, 6 hours for 4096 and they estimate 100+ hours for 8192 bit,
and 100,000 CPU hours for 16384 or about 3200 hours on the system that
crunched the other keys.

However, its also suggested that an effective attacks doesn't need to
know all the keys, since most keys would likely have been generated
within the first 3000 PIDs, which greatly reduce the number of
possibilities.

Hopefully anyone using an effected system has changed all their personal
keys + host keys + ssl certs by now, or implemented some kind of brute
force detection/blocking, cause if the bots haven't started any brute
force attacks yet, I'm guessing they will sooner rather then later.

-- 

Best regards,
 Duane

http://www.freeauth.org - Enterprise Two Factor Authentication
http://www.nodedb.com - Think globally, network locally
http://www.sydneywireless.com - Telecommunications Freedom
http://e164.org - Global Communication for the 21st Century

"In the long run the pessimist may be proved right,
    but the optimist has a better time on the trip."