[ClusterLabs] Transition aborted when disabling resource

Klaus Wenninger kwenning at redhat.com
Thu Sep 28 05:25:55 EDT 2017


On 09/28/2017 11:09 AM, Roberto Muñoz Gomez wrote:
>
>     > It is common to get a "Transition aborted" error when try to disable
>     > a resource?
>
>     Yes, "Transition aborted" is not an error (notice the log is at
>     "info:"
>     level), just an indication that something in the configuration or
>     status changed, and requires a new transition calculation.
>
>
>
> If its not an error, then why the resource is not disable afterwards?
> The command seems to be forgotten.

It is very well possible that there is an error.
The transition being aborted is not an error because a configuration
change would mean that pacemaker (pengine in particular) has to
reconsider the decisions made before. The calculated transition is
kind of a write-up was has to be done next.
The fact that the transition is being aborted makes it actually very
reasonable that the configuration had been updated.

Did you check the config (like  target-role) or did you just see the
service still running by other means (systemctl status, ps, ...)?

Possible errors that prevent a service from being stopped should
appear in the logs after pacemaker has calculated a new transition
and faces problems executing it (e.g. resource-agent having
problems to stop the resource).

Regards,
Klaus

>
> Regards.
> Roberto.
>
>
> *P*///Antes de imprimir, piensa en el MEDIO AMBIENTE/
> AVISO LEGAL/DISCLAIMER <http://www.bolsasymercados.es/disclaimer.html>
>
>
>
> _______________________________________________
> Users mailing list: Users at clusterlabs.org
> http://lists.clusterlabs.org/mailman/listinfo/users
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20170928/842e6c71/attachment-0003.html>


More information about the Users mailing list