[Pacemaker] long time to start

[Schaefer, Diane E]

>> Actually this RA waits for the sync to complete. If it takes longer than

>> the allotted time-out, Pacemaker SIGTERM/SIGKILLs it. The issue is if it

>> can never complete in the allotted time frame...



>Then make the timeout longer?



I could make the timeout longer, although I'm not sure that would help in all cases.  I am most concerned about a resource that can not start due to some db issue.  This seems to put my entire cluster on hold until the timeout and then the failure threshold is reached which may take quite a long time.  I would have expected that other monitor "not running" status returns could be serviced and restarted while this one thread is hung.  In my case, the resources are somewhat independent and I'd expect one hung resource not to stop other resources defined from restarting if necessary.



Diane Schaefer


THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20100419/bd373bc7/attachment-0002.htm>