[Linux-cluster] help help, nodes restarting each other through HP iLO fencing and vice versa

[Daniel EPEE LEA]

Dave,
It's at startup. I got the 
Both node hang at "Service fenced start".

So I restarted node1. Node2 proceeded correctly once
node 1 was restarting. Then when fenced started in
node1, node 2 was fenced, and rebooted. :(


let me go through what you adviced.

Much regards

Daniel

--- David Teigland <teigland at redhat.com> wrote:

> On Mon, Feb 06, 2006 at 01:14:30PM -0800, Daniel
> EPEE LEA wrote:
> > Hi Everyone,
> > 
> > Please find appended my cluster.conf file. I have
> 2
> > nodes (DL 580G2 + iLO + RHEL ES v4 + RHCS + GFS) +
> SAN
> > 
> > I am seting up a cluster, but when fenced stars,
> on
> > node1 (rs1.domain.com) first, node 2
> (rs2.domain.com)
> > restart, and vice versa.  can someone help me fix
> > this, please
> 
> Are the nodes fencing each other when they start up?
>  If so, the way to
> avoid this is to make sure both nodes have joined
> the cluster before
> either node runs 'fence_tool join' (which starts
> fenced).  Further
> explanation in fenced man page.
> 
> Dave
> 
> 


-----------------------------------------------------------------------------
T O    G O D       B E        T H E         G L O R Y     :)
------------------------------------------------------------------------------

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com