[Linux-cluster] CS4 Virtual Ip address Service re-starts

[Lon Hohberger]

On Wed, 2006-03-29 at 15:52 +0800, See Chit Wee wrote:
> Message Classification: Restricted
> 
> I'm using RHEL CS4 and the objective is to create a 2-node Cluster,
> IP: 10.10.10.35 and IP: 10.10.10.39, using Active/Passive
> configuration.  
>  
> For a start, I want to create a Virtual IP address (10.10.10.22)
> service that will failover to the other node when one of the nodes is
> down. After creation of the service called "mf", I enabled the service
> in the Cluster Configuration GUI and it is started okay. In the
> "var/log/messages" file, after about approx. 2 minutes, I get the
> following messages : 
>  
> clurgmgrd [7838]:<notice> status on ip "10.10.10.22" returned 1
> (generic error) 
> clurgmgrd [7838]:<notice> Stopping service mf 
> clurgmgrd [7838]:<notice> Service mf is recovering 
> clurgmgrd [7838]:<notice> Recovering failed service mf 
> clurgmgrd [7838]:<notice> Service mf started 
>  
> After about 2 minutes, the whole affair repeats itself again for
> eternity. 
>  
> Is there any script for the virtual ip address that I can access to
> set the status ?  

This is solved in current CVS - and will be solved with U3.  It's got to
do with the router-pinging code being broken.

To get rid of it for now, delete the depth="20" monitor/status actions
from /usr/share/cluster/ip.sh and restart rgmanager.

-- Lon

>