[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] GFS locks recovery

On Fri, Mar 30, 2007 at 12:02:23AM +0200, Christos Triantafillou wrote:
> >Yes.  Did the remaining node have quorum when you killed the other?  If
> >not, then you should set two_node=1 in cluster.conf so it will.  Fencing,
> >dlm recovery and gfs recovery won't happen unless there's quorum; after
> >this recovery, the locks you want should be granted (regardless of whether
> >the other node has rebooted or not).
> Is there a way to see whether a node has the quorum at any given time?
> Or whether GFS recovery has taken place?

'cman_tool status' will tell you about quorum.  But, if only one node in a
two node cluster is up, and you don't have two_node=1, then we already
know that your cluster didn't have quorum.

group_tool will tell you the recovery status of fencing, dlm and gfs.

> >Either, possibly; you'd have to try it out.  GFS works much better with
> >flock (although that's not interruptible either), if that's an option.
> How flock() works better than fcntl() on GFS?

Despite being used for similar things, they are completely different
commands with completely different implementations.  Since flock() is a
far simpler command, its implementation is far simpler, faster, and works
much better in general.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]