[Pacemaker] Contraining clones per node

[Jens.Braeuer at rohde-schwarz.com]

Hi everyone,

i have been into heartbeat 2 and pacemaker for some time now and wonder 
wheather i can use it in more than just the "normal" HA situation. However 
going through the excelent "Pacemaker Configuration Explained" or the 
Linux HA Cluster book by Michael Schwarzkopff, i still have no idea how to 
configure pacemaker for my scenario.

My environment consists of multiple servers (~40), each with one or more 
cpu-cores. I have two application-types called A and B (services like eg. 
apache), that each use one cpu core. A is mission critical, B is optional.
So what i want to express is that there should be 20 A's and the remaining 
cpu's may be used by B's. When a node executing A's fails, it is perfectly 
ok to shut down B's to make cpu cores available for A's to be started.

Any idea how to do this?

Going through the various examples in the book and pdf, i found examples 
on how to use instance-attributes for one resource. That mean things like 
"start apache only on host with more than XY ram or MN cpu speed". 
However, in my scenario i thing i need contraints that invole the number 
of resources on the host. An example would be "the sum of A's and B's 
started on node must be less or equal the number of cpu cores". But even 
going through the parameters supplied to ocf-agent (page 72 in the 
pacemaker explained pdf), it seem i am unable to figure out how many 
clones a currently runs.

Is pacemaker able to handle such constraints? Is there some work-around 
(eg with score-values) to emulate such behavior?

any ideas/hints/comments are very welcome.
best regards,

Jens Bräuer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091130/87a10ec7/attachment.html>