[Pacemaker] Trouble getting two node cluster to failover when network lost

Stefan Bauer stefan.bauer at cubewerk.de
Wed Mar 19 18:17:44 UTC 2014


If you dont care about the second nics status, why do you even use it redundant?

Stefan

> Am 19.03.2014 um 19:13 schrieb "Aaron Wilson" <awilson at nautilusgrp.com>:
> 
> Hello,
> 
> This is my first post and I am new to HA clusters.  Let me start by thanking everyone for their contributions to such a great project. 
> 
> I am really hoping to get pointed in the right direction so I can stop with the guessing.  
> 
> On Ubuntu 12.04 I am using corosync, cman and pacemaker I have two servers each with 2 nics. Both servers are identical, having each of their nics on the same respective subnets. 
> 
> When the master server fails completely, the Virtual IPs are taken over by the failover server (works as expected).  What I want is to failover when connectivity is lost on either of the nics, not just when the all communication to the master server fails.  
> 
> Not sure if this configuration is done on the Corosync as in Redundant Ring, or the pacemaker side.  I have tried several configurations with no difference.  I even tried using the RA ethmonitor, but it fails to start the monitor. 
> 
> Any advice would be greatly appreciated. 
> 
> Thanks You,
> 
> 
>   Aaron
> _______________________________________________
> 
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> 
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> 
> 
> 
> Project Home: http://www.clusterlabs.org
> 
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> 
> Bugs: http://bugs.clusterlabs.org
> 





More information about the Pacemaker mailing list