[Pacemaker] pacemaker 1.0.2 rsc_defaults section

Eliot Gable egable at broadvox.net
Wed Apr 29 13:17:45 UTC 2009


Andrew, thanks for the assistance. FYI, I got that bit of code from page 35 (PDF page 47) of your Configuration_Explained.pdf file for version 1.0.2: http://clusterlabs.org/mediawiki/images/f/fb/Configuration_Explained.pdf. You may want to fix that page.

Also, now it seems that the verify tool does not allow me to do colocation without a resource_set:

cib.xml:318: element rsc_colocation: Relax-NG validity error : Expecting an element resource_set, got nothing
cib.xml:318: element rsc_colocation: Relax-NG validity error : Element rsc_colocation failed to validate content

Page 24 (PDF page 36) shows an example of using rsc_colocation without the resource_set. Also, FYI, you have a couple places where you wrote collocation instead of colocation (which are not the same thing): PDF pages 35, 39, 40, 74. Also, with resource_set usage, it is no longer clear whether I should use rsc= or with-rsc= at all, much less where I would use them. And, on page 27-28 (PDF Pages 39-40), you write:

"The only thing that matters is that in order for any member of set N to be active, all the members of set N+1 must also be active (and naturally on the same node), and that if a set has sequential="true", then in order for member M to be active, member M+1 must also be active."

Do you mean N-1 and M-1 instead of N+1 and M+1? All of the other ordering mechanisms take a top-down approach to 'active' requirements, so why would these sets be bottom-up? For example, page 26 (PDF page 38) shows the exact opposite ordering.

In your example on page 28 (PDF Page 40), are you saying that G and F both must have Master state for C,D,E,B,A to be colocated with G and F? Or is it that one of them must be in Master state, or just G must be in Master state? I assume it is the first. In other words, G and F must be colocated in the Master state (Master/Slave or Slave/Master is not allowed) and that whichever node ends up with both of them in the Master state, that node also gets resources E,D,C and B,A. If that is the case, then how would you express that two multi-state resources are preferred to be in Master/Slave or Slave/Master colocation, but Master/Master is also allowed (just lower preference)? Would you assign role="Master" with a score="-1"?

Thanks again for your assistance.




Eliot Gable
Senior Engineer
1228 Euclid Ave, Suite 390
Cleveland, OH 44115

Direct: 216-373-4808
Fax: 216-373-4657
egable at broadvox.net


CONFIDENTIAL COMMUNICATION.  This e-mail and any files transmitted with it are confidential and are intended solely for the use of the individual or entity to whom it is addressed. If you are not the intended recipient, please call me immediately.  BROADVOX is a registered trademark of Broadvox, LLC.

-----Original Message-----
From: Andrew Beekhof [mailto:beekhof at gmail.com]
Sent: Wednesday, April 29, 2009 4:08 AM
To: pacemaker at oss.clusterlabs.org
Subject: Re: [Pacemaker] pacemaker 1.0.2 rsc_defaults section

On Wed, Apr 29, 2009 at 00:16, Eliot Gable <egable at broadvox.net> wrote:
> Pacemaker does not seem to like rsc_defaults:
>
>
>
> Apr 28 21:46:50 xxxxxxx cib: [3491]: ERROR: Extra element rsc_defaults in
> interleave
>
> Apr 28 21:46:50 xxxxxxx cib: [3491]: ERROR: Element configuration failed to
> validate content
>
> Apr 28 21:46:50 xxxxxxx cib: [3491]: ERROR: Element cib failed to validate
> content
>
> Apr 28 21:46:50 xxxxxxx cib: [3491]: ERROR: readCibXmlFile: CIB does not
> validate with pacemaker-1.0
>
>
>
> [root at xxxxx ~]# crm_verify -V -x cib.xml
>
> Relax-NG validity error : Extra element rsc_defaults in interleave
>
> cib.xml:10: element rsc_defaults: Relax-NG validity error : Element
> configuration failed to validate content
>
> cib.xml:3: element cib: Relax-NG validity error : Element cib failed to
> validate content
>
> crm_verify[3407]: 2009/04/28_22:07:05 ERROR: main: CIB did not pass
> DTD/schema validation
>
> Errors found during check: config not valid
>
>
>
> cib.xml:
>
>
>
> <?xml version="1.0"?>
>
> <!DOCTYPE cib SYSTEM "crm-1.0.dtd">
>
> <cib generated="true" admin_epoch="0" epoch="0" num_updates="0"
> have-quorum="false" validate-with="pacemaker-1.0">

remove the generated= attribute

>
>   <configuration>
>
>     <crm_config>
>
>       <cluster_property_set id="crm_config_prop-cluster-recheck">
>
>         <nvpair id="option-1" name="cluster-recheck-interval" value="15m"/>
>
>       </cluster_property_set>
>
>     </crm_config>
>
>     <rsc_defaults>
>
>       <meta_attributes id="off-hours" score="2">
>
>         <rule id="off-hour-rule">

rules need a score

>           <date_expression id="1amTO6am" operation="date_spec">

id's cant start with an integer

>             <date_spec hours="3-6" days="1-7"/>

needs an id=
days= -> weekdays=

>
>           </date_expression>
>
>         </rule>
>
>         <nvpair id="off-stickiness" name="resource-stickiness" value="0"/>
>
>       </meta_attributes>
>
>       <meta_attributes id="core-hours" score="1">
>
>         <nvpair id="core-stickiness" name="resource-stickiness"
> value="INFINITY"/>
>
>       </meta_attributes>
>
>     </rsc_defaults>
>
> ...snip...
>
>
>
>
>
> Any suggestions?
>
>
>
> Thanks in advance for the assistance.
>
>
>
> Eliot Gable
> Senior Engineer
> 1228 Euclid Ave, Suite 390
> Cleveland, OH 44115
>
> Direct: 216-373-4808
> Fax: 216-373-4657
> egable at broadvox.net
>
> CONFIDENTIAL COMMUNICATION.  This e-mail and any files transmitted with it
> are confidential and are intended solely for the use of the individual or
> entity to whom it is addressed. If you are not the intended recipient,
> please call me immediately.  BROADVOX is a registered trademark of Broadvox,
> LLC.
>
>
>
> ________________________________
> CONFIDENTIAL. This e-mail and any attached files are confidential and should
> be destroyed and/or returned if you are not the intended and proper
> recipient.
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
>

_______________________________________________
Pacemaker mailing list
Pacemaker at oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker

CONFIDENTIAL.  This e-mail and any attached files are confidential and should be destroyed and/or returned if you are not the intended and proper recipient.




More information about the Pacemaker mailing list