Red Hat Bugzilla – Bug 125582
Crash when rebooting inactive node
Last modified: 2009-04-16 16:14:59 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0;
T312461; .NET CLR 1.1.4322)
Description of problem:
On a two node, active passive system, re-boot of inactive node
crashes "active node". Active node reboots along with rebooting
system. There is nothing in the logs (cluster or otherwise) to
indicate any problem other then the reboot.
Version-Release number of selected component (if applicable):
Rh-Cluster 3.1 (update 1)
Steps to Reproduce:
1.shutdown -r now (on inactive node)
2.active node reboots...
Actual Results: Reboot of active node
Expected Results: No Action
As of U1, clumanager has not been QA'd on ia32e or x86_64 (though
packages are available).
I attempted to reproduce this on i386, but was unsuccessful. It may
or may not architecture dependent.
Hmm, have you tried the version from the beta channel?
I had forgotton all about this. My apologies. This can be closed. We
had been in the middle of re-iping our entire network. I had
neglected to updated the "Tie-Breaker" IP address. So as long as both
nodes were up and able to ping each other things were fine. However
if one node was taken down, remaining node assumed it was split brain
when it couldn't reach the "Tie-Breaker" and shut itself down.
This can be closed as a user error.
Fixing product name. Clumanager on RHEL3 was part of RHCS3, not RHEL3