[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] errors in SchedLog on submit host.



Dear condor team,
Please respond directly as I am not currently subscribed to the
condor_users list.

We are seeing these errors in the SchedLog on our one of our submit
nodes:

8/23 13:47:17 condor_write(): Socket closed when trying to write buffer
8/23 13:47:17 Buf::write(): condor_write() failed
8/23 13:47:17 SECMAN: Error sending response classad!
8/23 13:47:52 condor_write(): Socket closed when trying to write buffer
8/23 13:47:52 Buf::write(): condor_write() failed
8/23 13:47:52 Can't send job ad to mgr

We are not sure if this affects any job submissions - the jobs queued
are waiting on very busy nodes and the users do not have the best
priorities. there are 903 jobs. 10 are running and the rest are idle.
We don't see errors like the above on other submit nodes and are concerned.

Does the condor team have any input?

[root@cmsosgce log]# condor_version
$CondorVersion: 6.7.6 Mar 15 2005 $
$CondorPlatform: I386-LINUX_RH9 $

[root@cmsosgce log]# uname -a
Linux cmsosgce.fnal.gov 2.4.21-27.0.1.ELsmp #1 SMP Sat Dec 25 14:00:03 CST 2004 i686 i686 i386 GNU/Linux


we are running SLF303 on this node.

Thanks for any help you can give us.

Lisa Giacchetti
lisa@xxxxxxxx

USCMS T1 Support
FERMILAB