Multiple Bind Servers Stop Responding

Chavvon Smith chavvon at transoptics.com
Fri Oct 3 06:08:11 UTC 2003


I have a bind problem that maybe someone here has seen.

I have two bind servers, one on W2k, version 9.2.1, with all updates and one
on Red Hat 7.3, bind-9.2.0-8, with kernel-2.4.18-3.  The name servers are
delivering about 150 queries per second each and they suddenly stop
amswering request for no reason numerous times during the day. This happens
at no certain time intervals.  Sometimes 10 minutes, sometimes 10 hours.  If
I am local on the computer, DNS queries work fine.  They just stop working
externally for some reason.

This error shows up in the windows log when it crashes:
socket.c:1621: INSIST(!sock->pending_recv) failed

Sometimes it crashes and shows this error, and sometimes they just stop
responding while the service continues to run.  I've looked through the
"messages" logs and the only other logs I see are for lame delegations,
nothing serious.  Any idea why these servers would stop responding to
queries even though the service continues to run? They don't have anything
in common (except the network) and are both masters of different domains.

Regards,

CS



This email has been checked for viruses by http://SafemaiL.com




More information about the bind-users mailing list