[ClusterLabs] Problem with stonith and starting services
c.hernandez at medlabmg.com
Wed Jul 5 11:02:56 EDT 2017
>>> But you definitely shouldn't have a fencing-agent that claims to have fenced
>>> a node if it is not sure - rather the other way round if in doubt.
> True! Which is why I mentioned it to be dangerous.
> But your fencing-agent is even more dangerous ;-)
Well.. my startup fencing does, first, a "ssh $NODE reboot -f" and then another reboot using AWS API. All the times the node is reachable, the ssh reboot succeded. And the rest of the times, using AWS reboot works 100% of the times.
So.... I think it's not dangerous to always return OK on fencing script when I'm 100% (or 99.9999%) sure that fencing works ;)
Anyway, it's not the cause of this behavior. I'll keep trying... I'm also thinking on post this issue on the developers cluster list
More information about the Users