preeves1 at gmail.com
Fri Apr 28 14:34:28 UTC 2006
I finally figured out our errors. Im not sure why or how this happened
but it did. I began doing a tcpdump of the DNS transactions between
the BIND servers and the master Window servers. I imported the file
into ethereal and began compaing it to the errors in /var/log/messages
on the bind server. What I found at first confused me. From what I
could see, the BIND server was periodically requesting mulitiple copies
of the same IXFR. This would then cause the malformed transaction
errors. After thinking about it for a while i thought to myself that
it was acting like two versions of bind was running at the sametime.
So I decided to do a ps -ef | grep named on the system. Guess what,
two PID's for bind running! I decided to restart the server and then
checked named again and only one was running. Not one malformed error
in almost a day! So basically the two BIND processes were sharing and
attempting to update the same bind files.
Im really not sure how this happened but I will probably speak with our
systems group about it.
More information about the bind-users