[Pacemaker] Problems with corosync while forking processes during node startup.

[Lars Marowsky-Bree]

On 2013-02-21T08:18:00, Andrew Beekhof <andrew at beekhof.net> wrote:

> > That's not supported on SLE HA. (Because it breaks rolling upgrades from
> > previous versions; we'll move to this post-SLE 11.)
> >
> > This looks like a problem to raise a support request for.
> Good luck with that, there's a reason "ver: 1" was created :)

We can always start declaring it supported as an option, and then
customers can make their own choice regarding whether they want this
(and need to schedule a whole-cluster-down maintenance window for the
migration) or preserve rolling upgrade functionality.

Or we fix the corosync problem with forking from a multi-threaded
program.  ;-) We've never really had customer report problems with this
either, but I'm not sure why that is, honestly. I know the problem
theoretically exists, but it has never hit us.

But since this is an unsupported environment with self-compiled
binaries, it may be tough to argue ...


Regards,
    Lars

-- 
Architect Storage/HA
SUSE LINUX Products GmbH, GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 21284 (AG Nürnberg)
"Experience is the name everyone gives to their mistakes." -- Oscar Wilde