[Linux-cluster] Freeze with cluster-2.03.11

[Steven Whitehouse]

Hi,

On Fri, 2009-03-27 at 20:02 +0100, Kadlecsik Jozsef wrote:
> On Fri, 27 Mar 2009, Wendy Cheng wrote:
> 
> > ... [snip] ...
> > > Sigh. The pressure is mounting to fix the cluster at any cost, and nothing
> > > remained but to downgrade to
> > > cluster-2.01.00/openais-0.80.3 which would be just ridiculous.
> > 
> > I have doubts that GFS (i.e. GFS1) is tuned and well-maintained on newer 
> > versions of RHCS (as well as 2.6 based kernels). My impression is that 
> > GFS1 is supposed to be phased out starting from RHEL 5. So if you are 
> > running with GFS1, why downgrading RHCS is ridiculous ?
> 
> We'd need features added to recent 2.6 kernels (like read-only bindmount), 
> so the natural path was upgrading GFS1. However, as in the present state 
> our cluster is unstable, either we have to find the culprit or go back to 
> the proven version (and loosing the required new features).

Read only bind mounts have not been tested with gfs1 and they might very
well not work correctly, so be careful. Our general plan is to try and
introduce new features into GFS2 and to maintain GFS with its existing
feature set. Thats not to say that there will be no new features in GFS,
but just that we are trying in general to put the new stuff in GFS2.

Upgrading from GFS1 to GFS2 will always have to involve a shutdown in
cluster operations since the differing journalling schemes rule out a
node by node in place upgrade I'm afraid,

Steve.