[Pacemaker] Enable remote monitoring

Andrew Beekhof andrew at beekhof.net
Thu Nov 8 23:10:35 UTC 2012


On Thu, Nov 8, 2012 at 7:15 PM, Lars Marowsky-Bree <lmb at suse.com> wrote:
> On 2012-11-08T16:15:50, Andrew Beekhof <andrew at beekhof.net> wrote:
>
>> > And no, I'm not proposing that we allow overriding the
>> > class/provider/type tuple for start/stop ;-)
>> Did you consider having the VirtualDomain do the nagios redirect for
>> monitor operations?
>> If so, what was the drawback?
>
> The Xen agent has a call-out. Sucks to configure, because there is no
> GUI magic (i.e., with the metadata for configuring that etc). And, yes,
> that'd work, and look pretty much the same thing - except that
> "nagios:httpd" (for example) would read "params callout=nagios:httpd".
>
> The downside is the configuration syntax and that each RA that an admin
> wants to call a different monitor for needs to support it,

Wouldn't there be only a couple though? Just Xen and VirtualDomain I'd
have thought.

> and the UI
> needs to be independently taught about it as well, etc.

Well you need to teach it about something regardless.
Having GUIs look for callout=nagios:* is not really less magical than
setting a class and type for the <op/>
Just saying...

>> > Third, since the "start" of the base container may return before the
>> > guest is fully booted (to stick with the VM resource), we may need an
>> > additional timeout here. We *could* abuse start-delay (which might
>> > finally give it some legitimate use), but looping until we got the first
>> > success also appears attractive.
>> My concern there is that there needs to be a finite termination point
>> for the "its still bad" looping.
>> No better ideas yet though.
>
> The monitor will only ever start after the base is at least started, so
> we can loop for, say, thrice the timeout or something.

Wait, who are you thinking should do the looping?  PE or lrmd or the RA?

>> > The one downside here is that, unless we modify the PE or make the
>> > update to the CIB special somehow, the tools can't show that those
>> > ops/services aren't yet reporting healthy. But I think this trade-off is
>> > acceptable. And might be useful in other scenarios too.
>> You do or dont want to show them as unhealthy?  I'm not parsing this well.
>
> I'd like to show that, but it isn't so easy in this approach for the
> initial monitor.

ok

>
>
> 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
>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org




More information about the Pacemaker mailing list