[ClusterLabs] Antw: parallel execution of resources

Ulrich Windl Ulrich.Windl at rz.uni-regensburg.de
Wed Oct 7 14:34:38 UTC 2015


>>> Vijay Partha <vijaysarathy94 at gmail.com> schrieb am 07.10.2015 um 11:39 in
Nachricht
<CAMmMq+YDuPq3oCbooGQDy6Ays7Luz0hxeLAPE+_75C-re8Ocfw at mail.gmail.com>:
> hi,
> 
>     I have around 30 services running in each node of a 2 node cluster.
> when more than one resource fails pacemaker tries to restart the resources
> but does it in a sequential way. is it possible for pacemaker to start
> resources parallely?

Normally it does (while following ordering constraints).

> 
> are there situations where pacemaker can reboot a node automatically?  I
> was working on a node and suddenly got rebooted. I am having a thought that
> pacemaker could be the reason. is the node rebooted  because of pacemaker?

It's called "fencing". Maybe have a look at yur syslog or cluster logs....

Regards,
Ulrich

> 
> thanks and regards
> p.vijay








More information about the Users mailing list