[Linux-cluster] fencing with gfs

[Jakub Suchy]

Hello,
I believe that this is normal and is caused by a recent update of cman to
cman-2.0.84-2.el5_2.2 - everything waits until fencing is completed.

However, your fencing seems misconfigured, this should be a correct
syntax for manual fence:
        <method name="1">
          <device name="human nodename="node.example.com"/>
        </method>

Jakub


-- 
Jakub Suchý <jakub.suchy at enlogit.cz>
GSM: +420 - 777 817 949

Enlogit s.r.o, U Cukrovaru 509/4, 400 07 Ústí nad Labem
tel.: +420 - 474 745 159, fax: +420 - 474 745 160
e-mail: info at enlogit.cz, web: http://www.enlogit.cz

Energy & Logic in IT


Enrico Popp wrote:
> Hello,
> 
> i'm new in this list. I have to show no understanding for the fence
> daemon. Why should i use this ? My issue:
> 
> I have a cluster with two nodes. And i want to use the dbrd 8.0 or
> higher for a primary/primary state. There was a problem with fencing. 
> When the first node is down, the second node can not be rebootet by
> root, because the shell says: 
> fence_tool: waiting for cluster quorum
> 
> Here my cluster.conf
> <?xml version="1.0"?>
> <cluster name="mail-intern" config_version="1">
> <cman two_nodes="1" expected_votes="1"/>
>         <clusternodes>
> 	 <clusternode name="mail-i01" nodeid="1">
> 	   <fence>
> 	     <method name="human">
> 	       <device name="maili01"	ipaddr="10.1.0.1"/>
>              </method>
>            </fence>
>           </clusternode>
>          <clusternode	name="mail-i02"	nodeid="2">
> 	    <fence>
> 	      <method name="human">
> 	        <device	name="maili02"	ipaddr="10.1.0.2" />
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20081209/6160bfe9/attachment.sig>