[Linux-HA] How to prevent unexpected unmount (stop) of ordered clone Filesystem resource on alive node

[Andrew Beekhof]

Hi,

Sorry for the delay... christmas vacation etc...

I looked into this, it seems that the PE considers configurstoreclone  
to be stopping (which is both true and necessary for the correct  
notifications to be sent).
However this means that the ordering constraints cause the other two  
clones to be restarted (in a misguided attempt to recover).

Its going to take some time to fix unfortunately.
Until it is, perhaps adding score="0" to the ordering constraints will  
be sufficient for your purposes.

Andrew

On Jan 3, 2008, at 4:35 AM, Takekazu Okamoto wrote:

> Andrew-san,
>
>>>> Andrew Beekhof <beekhof at gmail.com> 2007/12/31 1:50 >>>
>> If you use hb_report I'll have a testcase which I can use to
> reproduce
>> the problem and be sure that the problem is 100% fixed.
>
> Attached is my the result of hb_report.
> This includes entire log.
> - Startup both node1 and node2
> - "ssh node1 pkill heartbeat" on node2 to generate failure
> - node2 tries to unmount (stop)
> - node2 reboots by itself.
>
> Thank you,
> Takekazu Okamoto
> <report1.tar.gz>_______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems