[Linux-cluster] GFS locks wnen a node fails

Benjamin Marzinski bmarzins at redhat.com
Thu Jul 26 18:24:50 UTC 2007


On Fri, Jul 20, 2007 at 01:39:13AM -0700, Hal wrote:
> Hallo everybody,
> I have a test cluster of 4 machines. node0 - gnbd server and gnbd fence server
> and 3 nodes to mount gfs. The problem is that when I unplug one of the nodes,
> gfs locks and no one can access it until the node is reconnected. 
> 
> How can this lock be avoided if one node fails? 
> How can I tell that gnbd-fencing is working at all?
> 
> "gnbd_import -c node0" says nothing even if I do "fence_node node2" I assume 
> fencing is not working am I right?

To check fencing, you can to a
# gnbd_import -s <node_to_fence> -t <server_to_fence_from>
# gnbd_import -c <server_you_fenced_from>

This will show the banned list on the server. The fenced node should be there.
The node will remain on the panned list until it is manually unfenced, or
until the gnbd module is reloaded, and it reimports devices from the server.
 
> Regards 
> Hal
> 
> 
>        
> ____________________________________________________________________________________Ready for the edge of your seat? 
> Check out tonight's top picks on Yahoo! TV. 
> http://tv.yahoo.com/
> 
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster




More information about the Linux-cluster mailing list