[Pacemaker] Problem with pingd.

Andrew Beekhof andrew at beekhof.net
Tue Feb 23 07:21:27 UTC 2010


http://www.clusterlabs.org/doc/en-US/Pacemaker/1.0/html/Pacemaker_Explained/ch-rules.html#id2083551

Look for the "type" field.

On Tue, Feb 23, 2010 at 5:08 AM, Jayakrishnan <jayakrishnanlll at gmail.com> wrote:
> Sir,
> I am afraid to ask you but how can I tell pacemaker to compare as number
> instead of string.
> I changed -inf: to -10000 in pingd location constarint but same problem
> persists.
> I also changer the global resource stickness to 10000. but still not
> working.
>
> With thanks,
> Jayakrishnan.L
>
> On Tue, Feb 23, 2010 at 1:04 AM, Andrew Beekhof <andrew at beekhof.net> wrote:
>>
>> On Mon, Feb 22, 2010 at 6:46 PM, Jayakrishnan <jayakrishnanlll at gmail.com>
>> wrote:
>> > Sir,
>> > I have setup a 2 node cluster with heartbeat 2.99  pacemaker 1.05. I am
>> > using Ubuntu 9.1. Both the packages are installed from ubuntu karmic
>> > repository.
>> > My packages are:
>> >
>> > heartbeat                   2.99.2+sles11r9-5ubuntu1
>> > heartbeat-common                     2.99.2+sles11r9-5ubuntu1
>> > heartbeat-common-dev                 2.99.2+sles11r9-5ubuntu1
>> > heartbeat-dev                        2.99.2+sles11r9-5ubuntu1
>> > libheartbeat2                        2.99.2+sles11r9-5ubuntu1
>> > libheartbeat2-dev                    2.99.2+sles11r9-5ubuntu1
>> > pacemaker-heartbeat                  1.0.5+hg20090813-0ubuntu4
>> > pacemaker-heartbeat-dev              1.0.5+hg20090813-0ubuntu4
>> >
>> > My ha.cf file, crm configuration are all attached in the mail.
>> >
>> > I am making a postgres database cluster with slony replication. eth1 is
>> > my
>> > heartbeat link, a cross over cable is connected between the servers in
>> > eth1.
>> > eth0 is my external network where my cluster IP get assigned.
>> > server1--> hostname node1
>> > node 1 192.168.10.129 eth1
>> > 192.168.1.1-->eth0
>> >
>> >
>> > servver2 --> hostname node2
>> > node2  192.168.10.130 eth1
>> > 192.168.1.2 --> eth0
>> >
>> > Now when I pull out my eth1 cable, I need to make a failover to the
>> > other
>> > node. For that i have configured pingd as follows. But it is not
>> > working. My
>> > resources are not at all starting when I give rule as
>> > rule -inf: not_defined pingd or pingd lte0
>>
>> You need to get 1.0.7 or tell pacemaker to do the comparison as a
>> number instead of as a string.
>>
>> >
>> > I tried changing the -inf: to inf: then the resources got started but
>> > resource failover is not taking place when i pull out the eth1 cable.
>> >
>> > Please check my configuration and kindly point out where I am missing.
>> > PLease see that I am using default resource stickness as INFINITY which
>> > is
>> > compulsory for slony replication.
>> >
>> > MY ha.cf file
>> > ------------------------------------------------------------------
>> >
>> > autojoin none
>> > keepalive 2
>> > deadtime 15
>> > warntime 10
>> > initdead 64
>> > initdead 64
>> > bcast eth1
>> > auto_failback off
>> > node node1
>> > node node2
>> > crm respawn
>> > use_logd yes
>> > ____________________________________________
>> >
>> > My crm configuration
>> >
>> > node $id="3952b93e-786c-47d4-8c2f-a882e3d3d105" node2 \
>> >         attributes standby="off"
>> > node $id="ac87f697-5b44-4720-a8af-12a6f2295930" node1 \
>> >         attributes standby="off"
>> > primitive pgsql lsb:postgresql-8.4 \
>> >         meta target-role="Started" resource-stickness="inherited" \
>> >         op monitor interval="15s" timeout="25s" on-fail="standby"
>> > primitive pingd ocf:pacemaker:pingd \
>> >         params name="pingd" hostlist="192.168.10.1 192.168.10.75" \
>> >         op monitor interval="15s" timeout="5s"
>> > primitive slony-fail lsb:slony_failover \
>> >         meta target-role="Started"
>> > primitive slony-fail2 lsb:slony_failover2 \
>> >         meta target-role="Started"
>> > primitive vir-ip ocf:heartbeat:IPaddr2 \
>> >         params ip="192.168.10.10" nic="eth0" cidr_netmask="24"
>> > broadcast="192.168.10.255" \
>> >         op monitor interval="15s" timeout="25s" on-fail="standby" \
>> >         meta target-role="Started"
>> > clone pgclone pgsql \
>> >         meta notify="true" globally-unique="false" interleave="true"
>> > target-role="Started"
>> > clone pingclone pingd \
>> >         meta globally-unique="false" clone-max="2" clone-node-max="1"
>> > location vir-ip-with-pingd vir-ip \
>> >         rule $id="vir-ip-with-pingd-rule" inf: not_defined pingd or
>> > pingd
>> > lte 0
>> > meta globally-unique="false" clone-max="2" clone-node-max="1"
>> > colocation ip-with-slony inf: slony-fail vir-ip
>> > colocation ip-with-slony2 inf: slony-fail2 vir-ip
>> > order ip-b4-slony2 inf: vir-ip slony-fail2
>> > order slony-b4-ip inf: vir-ip slony-fail
>> > property $id="cib-bootstrap-options" \
>> >         dc-version="1.0.5-3840e6b5a305ccb803d29b468556739e75532d56" \
>> >         cluster-infrastructure="Heartbeat" \
>> >         no-quorum-policy="ignore" \
>> >         stonith-enabled="false" \
>> >         last-lrm-refresh="1266851027"
>> > rsc_defaults $id="rsc-options" \
>> >         resource-stickiness="INFINITY"
>> >
>> > _____________________________________
>> >
>> > My crm status:
>> > __________________________
>> >
>> > crm(live)# status
>> >
>> >
>> > ============
>> > Last updated: Mon Feb 22 23:15:56 2010
>> > Stack: Heartbeat
>> > Current DC: node2 (3952b93e-786c-47d4-8c2f-a882e3d3d105) - partition
>> > with
>> > quorum
>> > Version: 1.0.5-3840e6b5a305ccb803d29b468556739e75532d56
>> > 2 Nodes configured, unknown expected votes
>> > 5 Resources configured.
>> > ============
>> >
>> > Online: [ node2 node1 ]
>> >
>> > Clone Set: pgclone
>> >     Started: [ node1 node2 ]
>> > Clone Set: pingclone
>> >     Started: [ node2 node1 ]
>> >
>> > ============================
>> >
>> > please help me out.
>> > --
>
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
>




More information about the Pacemaker mailing list