I have a 6 node cluster running SLES 10sp2 OES 2sp1 that nodes occasionally
restart for no apparent reason. Clustering is working fine so there is
rarely user impact but I need to get this fixed. /var/log/messages shows no
errors, just a gap during the restart time. I can't find any other logs
with relevant info in them, any suggestions for specific ones to check? The
hardware is all HP DL 380 G5 servers. I added apm=off and acpi=off to the
GRUB loader and that may have helped a bit but just last night one of the
nodes restarted itself again.

Any other ideas? Should I try SP3 and OES 2 sp2? Any suggestions if I go
that route? Things are otherwise fine, if I can solve this w/o patching
would be OK by me. Running services include GroupWise 8, DNS, DHCP, and
file access for users. NSS volumes.

Todd Bowman
Senior IS Professional
University of Minnesota Physicians