Re: GFS upgrade questions

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

 



On Thu, 18 Dec 2008, Kadlecsik Jozsef wrote:

On Thu, 18 Dec 2008, Fabio M. Di Nitto wrote:

We are planning to upgrade our stable - but relative old - GFS1 cluster to
the new version of the stable branch. However as this will be our first
GFS1 upgrade, it's pretty unknown to us and there are some questions about
it I'd like to ask:

Could you be more specific from what version of the software you want to
upgrade to what version?

- Is it possible to upgrade the machines one by one from GFS
 (cman/fencing/dlm/etc.) point of view?
- Is it possible to upgrade the machines one by one from the background
 dependencies (openais, clvmd) point of view?
- What about an upgrade from a non-corosync-dependent setup to a
 corosync-dependent one?
- Should we watch out any new cluster configuration option?

I'll be able to answer those questions in details, once you can be more
specific.

This is cluster-2.02.00 with openais-0.80.3.


So your best upgrade path is to get latest openais-0.80.X serie and cluster-2.03.xx.

You should be able to upgrade one node at a time without issues. If it doesn't work as expected it is a bug so please make sure to notify us ASAP.

Upgrade path to a corosync based cluster is not working yet because corosync is not on wire compatible with openais at this point in time. It will be before 1.0 release. Note that cluster-2.03.xx stable cannot run on top of corosync and that cluster-3.0.0.alpha1 is well.. alpha stage and can be buggy.

So to answer all your questions (assuming you will go 2.03.xx):

- Is it possible to upgrade the machines one by one from GFS
 (cman/fencing/dlm/etc.) point of view?

Yes.

- Is it possible to upgrade the machines one by one from the background
 dependencies (openais, clvmd) point of view?

Yes.

Assuming you are using a distribution such as Fedora, both will come down together. You will need to rebuild the GFS kernel module manually (depending on the distribution)

- What about an upgrade from a non-corosync-dependent setup to a
 corosync-dependent one?

No. This is not an option for you at the moment.

- Should we watch out any new cluster configuration option?

It is always a good idea to look at possible new options by looking at the changelogs.

Fabio

--
I'm going to make him an offer he can't refuse.

--
Linux-cluster mailing list
Linux-cluster@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/linux-cluster

[Index of Archives]     [Corosync Cluster Engine]     [GFS]     [Linux Virtualization]     [Centos Virtualization]     [Centos]     [Linux RAID]     [Fedora Users]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite Camping]

  Powered by Linux