0

On a couple of occassions jobs on my Redhat Linux server hang up due to IBM Director's "cimserver" processing using 100% of the CPU.

Seen on,

Redhat 2.6.18-8.el5PAE #1 SMP Redhat 2.6.9-67.ELsmp #1 SMP

This Technote (in Japanese!!) seems to deal with a similar problem:

www-06.ibm.com/jp/domino04/pc/support/Sylphd07.nsf/jtechinfo/SYJ0-0186A56

Does anyone know what triggers this?

Stuart Woodward
  • 1,343
  • 4
  • 14
  • 29
  • What version of Director? What release/patchlevel of RHE? – MikeyB Jun 01 '09 at 01:40
  • I'm not the administrator of Director on this sytem. Is there a simple way to query it from the command line to get this information? Redhat release numbers were given above. – Stuart Woodward Jun 01 '09 at 04:00

1 Answers1

1

Assuming this is a binary only program, the best way to work out what it's doing is using strace. The first thing to do is to find out what it's doing now:

strace -f -o /tmp/cimserver.log -p $process_id

You can then look in /tmp/cimserver.log to see what system calls it's doing. You may find that the problem happened before now, in which case you'll need to run the process under strace from the start.

strace -f -o /tmp/cimserver.log /usr/sbin/cimserver $args

Beware this will slow the process down a reasonable amount. You can reduce the logging to particular system calls by using the -e option. Check the man page for more info.

Hopefully this will give you some pointer to the problem. Of course it could just be a bug in the software :)

David Pashley
  • 23,497
  • 2
  • 46
  • 73