Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Linux-HA: Pacemaker

Failed Actions

 

 

Linux-HA pacemaker RSS feed   Index | Next | Previous | View Threaded


gregg at damagecontrolusa

May 4, 2012, 12:57 PM

Post #1 of 4 (595 views)
Permalink
Failed Actions

I have two drbd nodes and three nodes that run mostly virtual machines.
The virtual machines can't run on the drbd nodes and drbd can't run on
the visualization hosts. I have location rules and everything works
fine. But I get a long list of Failed Action in the crm_mon output
saying that things are not installed. This is mostly an annoyance but it
does make it difficult to spot real problems because you get used to
ignoring the output.

Is there a preferred method to eliminate this?

One other question is about the location rules. I really only have two
rules but have ended up creating multiple instances of the same rule for
each resource. Is there a way to consolidate the rules?

Thanks in advance.

_______________________________________________
Pacemaker mailing list: Pacemaker [at] oss
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


david at davidcoulson

May 4, 2012, 1:01 PM

Post #2 of 4 (537 views)
Permalink
Re: Failed Actions [In reply to]

Why not run two separate clusters - One for VMs, one for DRBD.

You can create a group containing the resources and have the location constraint reference the group - You probably want to set the group to 'ordered=false' and 'collocated=false'. That said, if you split your environment into two clusters you don't need that anyway.

David


On May 4, 2012, at 3:57 PM, Gregg Stock wrote:

> I have two drbd nodes and three nodes that run mostly virtual machines. The virtual machines can't run on the drbd nodes and drbd can't run on the visualization hosts. I have location rules and everything works fine. But I get a long list of Failed Action in the crm_mon output saying that things are not installed. This is mostly an annoyance but it does make it difficult to spot real problems because you get used to ignoring the output.
>
> Is there a preferred method to eliminate this?
>
> One other question is about the location rules. I really only have two rules but have ended up creating multiple instances of the same rule for each resource. Is there a way to consolidate the rules?
>
> Thanks in advance.
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker [at] oss
> 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


_______________________________________________
Pacemaker mailing list: Pacemaker [at] oss
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


gregg at damagecontrolusa

May 4, 2012, 2:27 PM

Post #3 of 4 (540 views)
Permalink
Re: Failed Actions [In reply to]

My thought on including the drbd nodes in the cluster was to avoid
having to disable the quorum and that it would help reduce the chance of
split brain problems. But this is my first cluster so I'm going on what
seems reasonable without any real experience.

The unordered, uncollocated group has been bashed a bit. But it would
seem to work.

Thanks,
Gregg

On 5/4/2012 1:01 PM, David Coulson wrote:
> Why not run two separate clusters - One for VMs, one for DRBD.
>
> You can create a group containing the resources and have the location constraint reference the group - You probably want to set the group to 'ordered=false' and 'collocated=false'. That said, if you split your environment into two clusters you don't need that anyway.
>
> David
>
>
> On May 4, 2012, at 3:57 PM, Gregg Stock wrote:
>
>> I have two drbd nodes and three nodes that run mostly virtual machines. The virtual machines can't run on the drbd nodes and drbd can't run on the visualization hosts. I have location rules and everything works fine. But I get a long list of Failed Action in the crm_mon output saying that things are not installed. This is mostly an annoyance but it does make it difficult to spot real problems because you get used to ignoring the output.
>>
>> Is there a preferred method to eliminate this?
>>
>> One other question is about the location rules. I really only have two rules but have ended up creating multiple instances of the same rule for each resource. Is there a way to consolidate the rules?
>>
>> Thanks in advance.
>>
>> _______________________________________________
>> Pacemaker mailing list: Pacemaker [at] oss
>> 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
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker [at] oss
> 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
>

_______________________________________________
Pacemaker mailing list: Pacemaker [at] oss
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


andrew at beekhof

May 9, 2012, 9:11 PM

Post #4 of 4 (521 views)
Permalink
Re: Failed Actions [In reply to]

On Sat, May 5, 2012 at 5:57 AM, Gregg Stock <gregg [at] damagecontrolusa> wrote:
> I have two drbd nodes and three nodes that run mostly virtual machines. The
> virtual machines can't run on the drbd nodes and drbd can't run on the
> visualization hosts. I have location rules and everything works fine. But I
> get a long list of Failed Action in the crm_mon output saying that things
> are not installed. This is mostly an annoyance but it does make it difficult
> to spot real problems because you get used to ignoring the output.
>
> Is there a preferred method to eliminate this?

Basically, add the RA. I think more recent versions are quieter about
this kind of thing.
Especially if you set symmetric-cluster=false (see Pacemaker Explained
for what that might mean for the rest of your config)

>
> One other question is about the location rules. I really only have two rules
> but have ended up creating multiple instances of the same rule for each
> resource. Is there a way to consolidate the rules?

Recent versions of 1.1.x can re-use the rule blocks in multiple constraints:
http://clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Pacemaker_Explained/s-reusing-config-elements.html

Does that help?

> Thanks in advance.
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker [at] oss
> 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

_______________________________________________
Pacemaker mailing list: Pacemaker [at] oss
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

Linux-HA pacemaker RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.