[Pacemaker] rsc_location does not work.

renayama19661014 at ybb.ne.jp renayama19661014 at ybb.ne.jp
Thu Dec 24 02:39:39 UTC 2009


Hi Dejan,

> The hb_report doesn't contain the latest pe input file, so one
> can't say if the two location constraints quoted above made
> the group move once in the past away from srv01.
I'm sorry. 
I attach a file of hb_report and pengine of the problem again.

> Could it be that due to high default-resource-stickiness
> (INFINITY) the group prefers to stay where it was once placed. At
> least in the configuration explained there is this stated in
> connection with ping attributes:
> 
> 	...  or instead you can tell the cluster only to prefer nodes
> 	with the most connectivity. Just be sure to set the
> 	multiplier to a value higher than that of resource-stickiness
> 	(and don’t set either of them to INFINITY).
> 
> The hb_report doesn't contain the latest pe input file, so one
> can't say if the two location constraints quoted above made
> the group move once in the past away from srv01.
> 
> Finally, are you sure you need these collocations:
> 
> colocation rsc_colocation02-1-1 inf: group02-1 clnPingd
> colocation rsc_colocation02-1-2 inf: group02-1 clnPingd2

I cannot understand your answer well.
I am because English is clumsy.

However, I tried the next setting.

 <nvpair id="rsc-default-meta_attributes-resource-stickiness" name="resource-stickiness" value="100"/>

The resource is arranged as expected when I make resource-stickiness 100 in the case of this time.

Will the meaning of your answer be this setting?

* I attached the file of hb_report and pengine of this setting(set resource-stickiness=100).

Best Regards,
Hideo Yamauchi.


--- Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:

> Hi Hideo-san,
> 
> On Tue, Dec 22, 2009 at 11:28:52AM +0900, renayama19661014 at ybb.ne.jp wrote:
> > Hi,
> > 
> > We constituted the complicated cluster of three nodes.(2ACT+1STB)
> > 
> > We built a cluster by the next combination.
> > 
> >  * corosync-1.1.2
> >  * Reusable-Cluster-Components-fa44a169d55f
> >  * Cluster-Resource-Agents-6f02f8ad7fd4
> >  * Pacemaker-1-0-d990c453b999
> > 
> > The resource of group02-1 hoped that it started in an active node.(srv01)
> 
> According to ptest and the current CIB it should be running on
> srv01:
> 
> [0]rondo:problem > ptest -s -x srv01/cib.xml | grep group02-1
> group_color: group02-1 allocation score on srv01: 200
> group_color: group02-1 allocation score on srv02: -1000000
> group_color: group02-1 allocation score on srv03: 100
> 
> There are two more location constraints:
> 
> location pingd-group02-1 group02-1 \
>         rule $id="pingd-grou02-1-rule-1" -inf: defined default_ping_set and default_ping_set lt
> 100
> location diskd-group02-1 group02-1 \
>         rule $id="diskd-grou02-1-rule-1" -inf: defined default_ping_set2 and default_ping_set2
> lt 100
> 
> Could it be that due to high default-resource-stickiness
> (INFINITY) the group prefers to stay where it was once placed. At
> least in the configuration explained there is this stated in
> connection with ping attributes:
> 
> 	...  or instead you can tell the cluster only to prefer nodes
> 	with the most connectivity. Just be sure to set the
> 	multiplier to a value higher than that of resource-stickiness
> 	(and don’t set either of them to INFINITY).
> 
> The hb_report doesn't contain the latest pe input file, so one
> can't say if the two location constraints quoted above made
> the group move once in the past away from srv01.
> 
> Finally, are you sure you need these collocations:
> 
> colocation rsc_colocation02-1-1 inf: group02-1 clnPingd
> colocation rsc_colocation02-1-2 inf: group02-1 clnPingd2
> 
> Thanks,
> 
> Dejan
> 
> > But, against rsc_location which I appointed, the resource is started in a standby node.(srv03)
> > 
> > ------ output crm_mon -------------------------
> >  Resource Group: UMgroup01
> >      UmDummy01  (ocf::heartbeat:Dummy01):       Started srv01
> >      UmDummy02  (ocf::heartbeat:Dummy): Started srv01
> >      UmIPaddr   (ocf::heartbeat:IPaddr):        Started srv01
> >  Resource Group: group02-1
> >      Dummy01-1  (ocf::heartbeat:Dummy): Started srv03
> >      Dummy01-2  (ocf::heartbeat:Dummy): Started srv03
> >  Resource Group: group02-2
> >      Dummy02-1  (ocf::heartbeat:Dummy): Started srv02
> >      Dummy02-2  (ocf::heartbeat:Dummy): Started srv02
> >  Resource Group: grpStonith1
> >      prmStonithN1-1     (stonith:external/ssh): Started srv03
> >  Resource Group: grpStonith2
> >      prmStonithN2-1     (stonith:external/ssh): Started srv01
> >  Resource Group: grpStonith3
> >      prmStonithN3-1     (stonith:external/ssh): Started srv02
> >  Clone Set: clnUMgroup01
> >      Started: [ srv01 srv03 ]
> >  Clone Set: clnPingd
> >      Started: [ srv01 srv02 srv03 ]
> >  Clone Set: clnPingd2
> >      Started: [ srv01 srv02 srv03 ]
> > ----------------------------------------------------
> > 
> > ------ cib.xml group02-1 rsc_location --------------
> >       <rsc_location id="grp02-1-1-location" rsc="group02-1" node="srv01" score="200"/>
> >       <rsc_location id="grp02-1-2-location" rsc="group02-1" node="srv03" score="100"/>
> >       <rsc_location id="grp02-1-3-location" rsc="group02-1" node="srv02" score="-INFINITY"/>
> > ----------------------------------------------------
> > 
> > Will my cib.xml be a mistake? 
> > Or will it be a bug?
> > 
> > And I attach hb_report.
> > 
> > Best Regards,
> > Hideo Yamauchi.
> > 
> > 
> 
> 
> > _______________________________________________
> > 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
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: probrem3-report.tar.bz2
Type: application/octet-stream
Size: 33442 bytes
Desc: 2413634437-probrem3-report.tar.bz2
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091224/9563ceba/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: problem3-pengine.zip
Type: application/x-zip-compressed
Size: 11524 bytes
Desc: 3789966502-problem3-pengine.zip
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091224/9563ceba/attachment-0008.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: non_problem-report.tar.bz2
Type: application/octet-stream
Size: 33966 bytes
Desc: 2483577759-non_problem-report.tar.bz2
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091224/9563ceba/attachment-0005.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: non_probrem-pengine.zip
Type: application/x-zip-compressed
Size: 61764 bytes
Desc: 2300972331-non_probrem-pengine.zip
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091224/9563ceba/attachment-0009.bin>


More information about the Pacemaker mailing list