[Pacemaker] why pacemaker does not control the resources

Andrey Groshev greenx at yandex.ru
Mon Nov 18 12:38:44 UTC 2013



15.11.2013, 03:19, "Andrew Beekhof" <andrew at beekhof.net>:
> On 14 Nov 2013, at 5:06 pm, Andrey Groshev <greenx at yandex.ru> wrote:
>
>>  14.11.2013, 02:22, "Andrew Beekhof" <andrew at beekhof.net>:
>>>  On 14 Nov 2013, at 6:13 am, Andrey Groshev <greenx at yandex.ru> wrote:
>>>>   13.11.2013, 03:22, "Andrew Beekhof" <andrew at beekhof.net>:
>>>>>   On 12 Nov 2013, at 4:42 pm, Andrey Groshev <greenx at yandex.ru> wrote:
>>>>>>    11.11.2013, 03:44, "Andrew Beekhof" <andrew at beekhof.net>:
>>>>>>>    On 8 Nov 2013, at 7:49 am, Andrey Groshev <greenx at yandex.ru> wrote:
>>>>>>>>     Hi, PPL!
>>>>>>>>     I need help. I do not understand... Why has stopped working.
>>>>>>>>     This configuration work on other cluster, but on corosync1.
>>>>>>>>
>>>>>>>>     So... cluster postgres with master/slave.
>>>>>>>>     Classic config as in wiki.
>>>>>>>>     I build cluster, start, he is working.
>>>>>>>>     Next I kill postgres on Master with 6 signal, as if "disk space left"
>>>>>>>>
>>>>>>>>     # pkill -6 postgres
>>>>>>>>     # ps axuww|grep postgres
>>>>>>>>     root      9032  0.0  0.1 103236   860 pts/0    S+   00:37   0:00 grep postgres
>>>>>>>>
>>>>>>>>     PostgreSQL die, But crm_mon shows that the master is still running.
>>>>>>>>
>>>>>>>>     Last updated: Fri Nov  8 00:42:08 2013
>>>>>>>>     Last change: Fri Nov  8 00:37:05 2013 via crm_attribute on dev-cluster2-node4
>>>>>>>>     Stack: corosync
>>>>>>>>     Current DC: dev-cluster2-node4 (172793107) - partition with quorum
>>>>>>>>     Version: 1.1.10-1.el6-368c726
>>>>>>>>     3 Nodes configured
>>>>>>>>     7 Resources configured
>>>>>>>>
>>>>>>>>     Node dev-cluster2-node2 (172793105): online
>>>>>>>>            pingCheck       (ocf::pacemaker:ping):  Started
>>>>>>>>            pgsql   (ocf::heartbeat:pgsql): Started
>>>>>>>>     Node dev-cluster2-node3 (172793106): online
>>>>>>>>            pingCheck       (ocf::pacemaker:ping):  Started
>>>>>>>>            pgsql   (ocf::heartbeat:pgsql): Started
>>>>>>>>     Node dev-cluster2-node4 (172793107): online
>>>>>>>>            pgsql   (ocf::heartbeat:pgsql): Master
>>>>>>>>            pingCheck       (ocf::pacemaker:ping):  Started
>>>>>>>>            VirtualIP       (ocf::heartbeat:IPaddr2):       Started
>>>>>>>>
>>>>>>>>     Node Attributes:
>>>>>>>>     * Node dev-cluster2-node2:
>>>>>>>>        + default_ping_set                  : 100
>>>>>>>>        + master-pgsql                      : -INFINITY
>>>>>>>>        + pgsql-data-status                 : STREAMING|ASYNC
>>>>>>>>        + pgsql-status                      : HS:async
>>>>>>>>     * Node dev-cluster2-node3:
>>>>>>>>        + default_ping_set                  : 100
>>>>>>>>        + master-pgsql                      : -INFINITY
>>>>>>>>        + pgsql-data-status                 : STREAMING|ASYNC
>>>>>>>>        + pgsql-status                      : HS:async
>>>>>>>>     * Node dev-cluster2-node4:
>>>>>>>>        + default_ping_set                  : 100
>>>>>>>>        + master-pgsql                      : 1000
>>>>>>>>        + pgsql-data-status                 : LATEST
>>>>>>>>        + pgsql-master-baseline             : 0000000002000078
>>>>>>>>        + pgsql-status                      : PRI
>>>>>>>>
>>>>>>>>     Migration summary:
>>>>>>>>     * Node dev-cluster2-node4:
>>>>>>>>     * Node dev-cluster2-node2:
>>>>>>>>     * Node dev-cluster2-node3:
>>>>>>>>
>>>>>>>>     Tickets:
>>>>>>>>
>>>>>>>>     CONFIG:
>>>>>>>>     node $id="172793105" dev-cluster2-node2. \
>>>>>>>>            attributes pgsql-data-status="STREAMING|ASYNC" standby="false"
>>>>>>>>     node $id="172793106" dev-cluster2-node3. \
>>>>>>>>            attributes pgsql-data-status="STREAMING|ASYNC" standby="false"
>>>>>>>>     node $id="172793107" dev-cluster2-node4. \
>>>>>>>>            attributes pgsql-data-status="LATEST"
>>>>>>>>     primitive VirtualIP ocf:heartbeat:IPaddr2 \
>>>>>>>>            params ip="10.76.157.194" \
>>>>>>>>            op start interval="0" timeout="60s" on-fail="stop" \
>>>>>>>>            op monitor interval="10s" timeout="60s" on-fail="restart" \
>>>>>>>>            op stop interval="0" timeout="60s" on-fail="block"
>>>>>>>>     primitive pgsql ocf:heartbeat:pgsql \
>>>>>>>>            params pgctl="/usr/pgsql-9.1/bin/pg_ctl" psql="/usr/pgsql-9.1/bin/psql" pgdata="/var/lib/pgsql/9.1/data" tmpdir="/tmp/pg" start_opt="-p 5432" logfile="/var/lib/pgsql/9.1//pgstartup.log" rep_mode="async" node_list=" dev-cluster2-node2. dev-cluster2-node3. dev-cluster2-node4. " restore_command="gzip -cd /var/backup/pitr/dev-cluster2-master#5432/xlog/%f.gz > %p" primary_conninfo_opt="keepalives_idle=60 keepalives_interval=5 keepalives_count=5" master_ip="10.76.157.194" \
>>>>>>>>            op start interval="0" timeout="60s" on-fail="restart" \
>>>>>>>>            op monitor interval="5s" timeout="61s" on-fail="restart" \
>>>>>>>>            op monitor interval="1s" role="Master" timeout="62s" on-fail="restart" \
>>>>>>>>            op promote interval="0" timeout="63s" on-fail="restart" \
>>>>>>>>            op demote interval="0" timeout="64s" on-fail="stop" \
>>>>>>>>            op stop interval="0" timeout="65s" on-fail="block" \
>>>>>>>>            op notify interval="0" timeout="66s"
>>>>>>>>     primitive pingCheck ocf:pacemaker:ping \
>>>>>>>>            params name="default_ping_set" host_list="10.76.156.1" multiplier="100" \
>>>>>>>>            op start interval="0" timeout="60s" on-fail="restart" \
>>>>>>>>            op monitor interval="10s" timeout="60s" on-fail="restart" \
>>>>>>>>            op stop interval="0" timeout="60s" on-fail="ignore"
>>>>>>>>     ms msPostgresql pgsql \
>>>>>>>>            meta master-max="1" master-node-max="1" clone-node-max="1" notify="true" target-role="Master" clone-max="3"
>>>>>>>>     clone clnPingCheck pingCheck \
>>>>>>>>            meta clone-max="3"
>>>>>>>>     location l0_DontRunPgIfNotPingGW msPostgresql \
>>>>>>>>            rule $id="l0_DontRunPgIfNotPingGW-rule" -inf: not_defined default_ping_set or default_ping_set lt 100
>>>>>>>>     colocation r0_StartPgIfPingGW inf: msPostgresql clnPingCheck
>>>>>>>>     colocation r1_MastersGroup inf: VirtualIP msPostgresql:Master
>>>>>>>>     order rsc_order-1 0: clnPingCheck msPostgresql
>>>>>>>>     order rsc_order-2 0: msPostgresql:promote VirtualIP:start symmetrical=false
>>>>>>>>     order rsc_order-3 0: msPostgresql:demote VirtualIP:stop symmetrical=false
>>>>>>>>     property $id="cib-bootstrap-options" \
>>>>>>>>            dc-version="1.1.10-1.el6-368c726" \
>>>>>>>>            cluster-infrastructure="corosync" \
>>>>>>>>            stonith-enabled="false" \
>>>>>>>>            no-quorum-policy="stop"
>>>>>>>>     rsc_defaults $id="rsc-options" \
>>>>>>>>            resource-stickiness="INFINITY" \
>>>>>>>>            migration-threshold="1"
>>>>>>>>
>>>>>>>>     Tell me where to look - why does pacemaker not work?
>>>>>>>    You might want to follow some of the steps at:
>>>>>>>
>>>>>>>       http://blog.clusterlabs.org/blog/2013/debugging-pacemaker/
>>>>>>>
>>>>>>>    under the heading "Resource-level failures".
>>>>>>    Yes. Thank you.
>>>>>>    I've seen this article and now I study it in more detail.
>>>>>>    A lot of information in the logs, so it is difficult to determine where the error is, and where the consequence of error.
>>>>>>    Now I'm trying to figure it out.
>>>>>>
>>>>>>    BUT...
>>>>>>    While I can say with certainty that the RA with monitor in the MS(pgsql) is called ONLY on the node on which the last was launched PACEMAKER.
>>>>>   It looks like you're hitting https://github.com/beekhof/pacemaker/commit/58962338
>>>>>   Since you appear to be on rhel6 (or a clone of rhel6), can I suggest you use the 1.1.10 packages that come with 6.4?
>>>>>   They include the above patch.
>>>>   I already use (builded from source two weeks ago)
>>>  Upstream 1.1.10 does not include the above patch.
>>  Strangely, in my source code, these lines exist.
>>  Maybe I do not collect build.
>>  I have the source - "master", and build RPM - 1.1.10.
>>>>   * pacemaker 1.1.10
>>>>   * resource-agents 3.9.5
>>>>   * corosync 2.3.2
>>>>   * libqb 0.16
>>>>   & CentOS 6.4
>>>>
>>>>   The same config work on pacemaker 1.1.9/corosync 1.4.5
>>>>   Not ideal, but no such problem.
>>>>
>>>>   At first idea - I thought I should move the target-role=Master from MS to primitive pgsql.
>>>>   And so even working.
>>>>   But after a crash killing the main PostgreSQL process - started the same.
>>>>   Today's experiments showed that this behavior starts after I add in MS "notify=true".
>>>>   But primitive pgsql not properly work without "notify" messages.
>>>>   While I in frustration :(
>>>>>   Also, just to be sure. Are you expecting monitor operations to detect when you started a resource manually?
>>>>>   If so, you'll need a monitor operation with role=Stopped. We don't do that by default.
>>>>   I expect that the resource monitoring on all the time, otherwise how to control them?
>>>  When a node joins the cluster, we check to see if it had any resources running.
>>>  If no-one has the resource running, we pick a node and start it there.
>>>
>>>  If a malicious admin then starts the resource somewhere else, manually, we would not normally detect this.
>>>  It is assumed that someone trusted with root privileges would not do this on purpose.
>>>
>>>  However, if you do not trust your admins, as explained above, you can configure pacemaker to periodically re-check the node to detect and recover from this situation.
>>  This individual servers allocated for development.
>
> Then you likely have a random version from Git.
> Thats probably not a great idea.

Ok, build 1.1.11 - works as expected!


>
>>  Their only one malicious admin - It's I.
>>  But, I trust myself yet. :)
>>
>>  And even if I run something "hands" , then each time the utility deployment of cluster:
>>  * Certainly removes all the packages that were installed prior to this (referring to the cluster (libqb, corosync1 / 2 , cman, pacemaker, resource-agents, pcs, crmsh, cluster-glue));
>>  * Deletes all files that might remain from a previous installation ( and configs , and temp) ;
>>  * Install my RPM's;
>>  * Turns off and disables all of the services that may interfere with the operation of the cluster ;
>>  * Cleared the old configuration, calculates new configs corosync and pacemaker;
>>  * Writes and runs their cluster alternately every synchronizing each node.
>>
>>  In this situation it is difficult to survive something forgotten.
>>  I reinstalled the OS would be if I had access to a level above :)
>>>>   Or I do not quite understand the question.
>>>>>>>    'crm_mon -o' might be a good source of information too.
>>>>>>    Therefore, I see that my resources allegedly functioning normally.
>>>>>>
>>>>>>    # crm_mon -o1
>>>>>>    Last updated: Tue Nov 12 09:27:16 2013
>>>>>>    Last change: Tue Nov 12 00:08:35 2013 via crm_attribute on dev-cluster2-node2
>>>>>>    Stack: corosync
>>>>>>    Current DC: dev-cluster2-node2 (172793105) - partition with quorum
>>>>>>    Version: 1.1.10-1.el6-368c726
>>>>>>    3 Nodes configured
>>>>>>    337 Resources configured
>>>>>>
>>>>>>    Online: [ dev-cluster2-node2 dev-cluster2-node3 dev-cluster2-node4 ]
>>>>>>
>>>>>>    Clone Set: clonePing [pingCheck]
>>>>>>        Started: [ dev-cluster2-node2 dev-cluster2-node3 dev-cluster2-node4 ]
>>>>>>    Master/Slave Set: msPgsql [pgsql]
>>>>>>        Masters: [ dev-cluster2-node2 ]
>>>>>>        Slaves: [ dev-cluster2-node3 dev-cluster2-node4 ]
>>>>>>    VirtualIP      (ocf::heartbeat:IPaddr2):       Started dev-cluster2-node2
>>>>>>
>>>>>>    Operations:
>>>>>>    * Node dev-cluster2-node2:
>>>>>>      pingCheck: migration-threshold=1
>>>>>>       + (20) start: rc=0 (ok)
>>>>>>       + (23) monitor: interval=10000ms rc=0 (ok)
>>>>>>      pgsql: migration-threshold=1
>>>>>>       + (41) promote: rc=0 (ok)
>>>>>>       + (87) monitor: interval=1000ms rc=8 (master)
>>>>>>      VirtualIP: migration-threshold=1
>>>>>>       + (49) start: rc=0 (ok)
>>>>>>       + (52) monitor: interval=10000ms rc=0 (ok)
>>>>>>    * Node dev-cluster2-node3:
>>>>>>      pingCheck: migration-threshold=1
>>>>>>       + (20) start: rc=0 (ok)
>>>>>>       + (23) monitor: interval=10000ms rc=0 (ok)
>>>>>>      pgsql: migration-threshold=1
>>>>>>       + (26) start: rc=0 (ok)
>>>>>>       + (32) monitor: interval=10000ms rc=0 (ok)
>>>>>>    * Node dev-cluster2-node4:
>>>>>>      pingCheck: migration-threshold=1
>>>>>>       + (20) start: rc=0 (ok)
>>>>>>       + (23) monitor: interval=10000ms rc=0 (ok)
>>>>>>      pgsql: migration-threshold=1
>>>>>>       + (26) start: rc=0 (ok)
>>>>>>       + (32) monitor: interval=10000ms rc=0 (ok)
>>>>>>
>>>>>>    In reality now killed (signal 4|6) the PG master and the penultimate slave PG.
>>>>>>    IMHO, even if I have something configured incorrectly, the inability to monitor the resource must cause a fatal error.
>>>>>>    Or is there a reason not to do so?
>>>>>>
>>>>>>    _______________________________________________
>>>>>>    Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>>>>>>    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.clusterlabs.org
>>>>>   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.clusterlabs.org
>>>>   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.clusterlabs.org
>>>  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.clusterlabs.org
>>  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.clusterlabs.org
> 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




More information about the Pacemaker mailing list