When this message happens name service times out.
Cinense, Mark
macinen at sandia.gov
Tue May 9 16:12:08 UTC 2000
Greetings all,
About a month ago we upgraded our 10mb NIC to a 100mb NIC on a SPARC
5 270mhz with 96mb of memory. Our environment is about 8500 + machines,
that includes the servers. I have a script that runs daily via cron, and
its job is to gather statistics on the nameserver. This script also
restarts named by getting the named.pid info, and doing a kill -ILL on that
pid. Well after the upgrade of the NIC's, I am now getting this message in
my message log.
May 9 07:04:40 ns4 named[22432]: There may be a name server already running
on
[127.0.0.1].53
May 9 07:04:40 ns4 named[22432]: There may be a name server already running
on
[127.0.0.1].53
May 9 07:04:40 ns4 named[22432]: deleting interface [127.0.0.1].53
May 9 07:06:40 ns4 named[22432]: There may be a name server already running
on
[134.253.16.5].53
May 9 07:06:40 ns4 named[22432]: There may be a name server already running
on
[134.253.16.5].53
May 9 07:06:40 ns4 named[22432]: deleting interface [134.253.16.5].53
May 9 07:06:40 ns4 named[22432]: not listening on any interfaces
May 9 07:06:40 ns4 named[22432]: Forwarding source address is [0.0.0.0].53
May 9 07:06:40 ns4 named[22455]: Ready to answer queries.
May 9 07:07:11 ns4 named[22455]: refused query on non-query socket from
[134.253.93.44].2072
May 9 07:07:11 ns4 named[22455]: refused query on non-query socket from
[134.253.22.3].53
So what I did is change the schedule to every other day instead of
everyday. Unfortunately this only fixed this for a week. Today it happened
again. Has anyone seen this error message? This is killing my uptime, and
I am wondering if we need new machines. By the way the cpu utilization
looks like this through top.
CPU states: 92.7% idle, 3.4% user, 4.0% kernel, 0.0% iowait, 0.0% swap
Memory: 96M real, 10M free, 62M swap free
PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU
COMMAND
22455 root 1 58 0 11M 9400K
sleep 14:05 3.85% named
thanks for any Reponses,
Mark A Cinense
More information about the bind-users
mailing list