Linux instability with bind 8.2.2p5

[Eric Johansson]

I have run into an interesting bit of instability with Red Hat 6.1 and bind
8.2.2p5.

We installed the basic Red Hat 6.1 system and then installed all of the
update RPMs that were current including the syslog and the bind updates.
I modified named.conf and suslog.conf to send log information to the console
in addition to the usual log file location.  I am now seeing the systems
lockup to the point where they need someone to hit the reset switch.  As
near as I can figure, the lockup is occurring in the syslog facility or in
the IPC leading to syslog.  The reason I say this is that any activity that
requires logging (new logins, sudo, etc.) are blocked from executing and
never time out.  It appears that only bind generates enough syslog activity
to trigger this lockup condition.

I've seen some discussion in the archives about this but I didn't find any
resolutions.  I'm going to play with different logging configurations to see
if I can isolate the problem.  I would appreciate hearing from people that
may have encountered and solved this problem before.

--- eric