[Pacemaker] questions about expected behaviour stonith:meatware

imnotpc imnotpc at rock3d.net
Thu Jun 16 11:32:30 UTC 2011


On Thursday, June 16, 2011 02:38:51 Florian Haas wrote:
> On 06/16/2011 12:50 AM, imnotpc wrote:
> >> Funny but it looks fairly unequivocal to me.
> > 
> > Yes and no. The message is clear but unless you have someone sitting at
> > a console 24/7 running tail on the log file, it has little value.
> > According to the ClusterLabs stonith docs (which I just realized you
> 
> > wrote, haha):
> Meatware requires operator intervention, that much is a given.
> _Notifying_ an operator that intervention is necessary, beyond logging
> to the console or a log file, is beyond meatware's domain.
> 
> However, it's extremely easy to combine meatware with automated
> monitoring: any time a node is meatware-fencing another, it creates a
> socket file named /var/run/meatware.<hostname>, where hostname is the
> name of the node being fenced. Configure your automated monitoring to
> page an operator any time this file is present, and you've solved the
> notification problem.
> 
> Hope this is useful.

It was. I had seen mention of that file when I'd run meatclient on a node that 
wasn't DC. That's actually the best idea so far as long as you only want 
notification of fencing since it only exists on the DC node. Well, in theory. 
It turns out I had a few stale files that hadn't been deleted and it seems that 
was the cause of those warnings and errors in my log. I still don't get a 
console message but that is of limited value anyway.

Thanks, Jeff




More information about the Pacemaker mailing list