On 10/25/2010 10:00 AM, Scholte van Mast, Rene wrote:
> Hello,
> I have some trouble with my heartbeat configuration.
>
> Everytime I want one node down for maintenance I run "/etc/ha.d/heartbeat
> standby" on the masters-system to move all resources to the other node.
> Resource takeover completes successfully. If the maintenance requires a
> reboot, the Script "/etc/rc2.d/S75heartbeat stop" executes automatically
> during reboot. But here is the problem. Stopping the heartbeat services
> executes the scripts included in the haressources-file a second time one the
> master system.
>
> Why does stopping heartbeat services on slave, re-run the
> haressources-scripts on the master server, although all resources are already
> running on master. Why heatbeat doesn't notice that all ressources are
> already running on the master system.
The old R1-style heartbeat code does this for two reasons:
1) The LSB permits it to do this harmlessly. It is never an error
to start an init script which is already started.
2) It closed a hole where it might have been moving a resource
around at the time a system crashed
The LSB says the following things are not errors:
starting an init script for a service which is already running
stopping an init script for a service which is already stopped
If your scripts don't permit that, according to the LSB (and heartbeat
and Pacemaker), they are broken.
-- Alan Robertson
[email protected]
--
Alan Robertson<[email protected]>
"Openness is the foundation and preservative of friendship... Let me claim
from you at all times your undisguised opinions." - William Wilberforce
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
