[ClusterLabs] Antw: No Cluster fun (split brain)

[Ulrich Windl]

Hi!

I just wanted to let you know about the recent events: There was a situation when all nodes were up and the network was fine and the SBD was operational
AND every node thought it is domain controller.

As a result the filesystems in our virtual machine images were severely corrupted (call ist desaster).
As a side effect, some recent configuration changes were last as the configuration was replaced with some older configuration (by the cluster)...

Regards,
Ulrich

>>> Ulrich Windl schrieb am 19.05.2015 um 17:50 in Nachricht <555B5BC0.ADC : 161 :
60728>:
> Hi!
> 
> I just wanted to tell you that two nodes in out three-node cluster (SLES11 
> SP3) went mad when the thrird node was cleanly rebooted (i.e. after rcopenais 
> stop). Going mad means both nodes built up a "retransmit list" and decided to 
> be DC for the cluster. When the third node came back, the communication 
> problems went away, but the cluster was unable to continue running resources. 
> It needed a reboot of one of the mad nodes and later a cleanup of resources 
> for the other mad node.
[...]