On 10/25/2013 02:53 PM, Joel Young wrote:> Folks,
>
> I'm running gluster on fedora 19 with a four node gluster peer group
> and two distributed filesystems. One replicated and one distributed
> replicated.
>
> What is the best procedure to do a yum upgrade to 1.4.1 from 1.4.0?
> Do I need to make sure that each client remounts?
>
> Do I need to stop my whole cluster? Do I need to kick off all my
> users? One of the filesystems is /home on the cluster.
>
> Do the rpm scripts do the right thing?
Yes they do.
RPMs for 3.4.1 are in the Fedora YUM Updates (stable) repository. Just
do a `yum update` or `yum update gluster\*` on your Fedora boxes.
IIRC some people suggest that you update the clients first, then do a
"rolling update" of your servers. I.e. update one server at a time by
stopping gluster on that server, update, then restart. Then do the next
server, and so on.
There are documentation topics on this and much more on
http://www.gluster.org.
--
Kaleb