[Pacemaker] Trouble with "Failed application of an update diff"
Andrew Beekhof
andrew at beekhof.net
Fri Jun 27 02:08:44 EDT 2014
On 10 Jun 2014, at 10:44 pm, Виталий Туровец <corebug at corebug.net> wrote:
> Hello there again!
> Here you are: http://pastebin.com/bUaNQHs1
> It's also identical on both nodes.
> Thank you!
>
>
> 2014-06-10 3:20 GMT+03:00 Andrew Beekhof <andrew at beekhof.net>:
>
> On 9 Jun 2014, at 11:01 pm, Виталий Туровец <corebug at corebug.net> wrote:
>
> > Hello there again, people!
> >
> > After upgrading both nodes to such SW versions:
> >
> > pacemaker.x86_64 1.1.10-14.el6_5.3
> > pacemaker-cli.x86_64 1.1.10-14.el6_5.3
> > pacemaker-cluster-libs.x86_64
> > pacemaker-libs.x86_64 1.1.10-14.el6_5.3
> > corosync.x86_64 1.4.1-17.el6_5.1 @updates
> > corosynclib.x86_64 1.4.1-17.el6_5.1 @updates
> >
> > I am still facing the same problem: slave in Master/Slave set of MySQL won't start.
> > Master actually works correctly.
> > Output of cibadmin -Q on both nodes is identical.
> >
> > And here's the log of what happens when i try to do "cleanup MySQL_MasterSlave": http://pastebin.com/J90NuyEX.
> > By now i have MySQL slave running in manual mode, but this definitely is not what i'm trying to achieve using Pacemaker.
> > Can anyone help with this?
Um, I see:
location cli-standby-MySQL_MasterSlave MySQL_MasterSlave \
rule $id="cli-standby-rule-MySQL_MasterSlave" -inf: #uname eq wb-db1
which tells pacemaker that the MySQL_MasterSlave resource isn't allowed on wb-db1.
Thats why only one instance is being started and promoted to master.
> > Again, my pacemaker configuration:
>
> Can you provide the 'cibadmin -Ql' output instead?
> We need the status section in order to comment.
>
> >
> > node wb-db1 \
> > attributes standby=off
> > node wb-db2 \
> > attributes standby=off
> > primitive ClusterIP IPaddr2 \
> > params ip=10.0.1.68 cidr_netmask=32 nic=bond0.100 \
> > op monitor interval=30s \
> > meta target-role=Started
> > primitive MySQL mysql \
> > params binary="/usr/bin/mysqld_safe" enable_creation=1 replication_user=slave_user replication_passwd=here_goes_the_password datadir="/var/lib/mysql/db" socket="/var/run/mysqld/mysqld.sock" config="/etc/my.cnf" reader_attribute=readerOK evict_outdated_slaves=false max_slave_lag=600 \
> > op monitor interval=30s \
> > op monitor interval=35s role=Master OCF_CHECK_LEVEL=1 \
> > op monitor interval=60s role=Slave timeout=60s OCF_CHECK_LEVEL=1 \
> > op notify interval=0 timeout=90 \
> > op start interval=0 timeout=120 \
> > op stop interval=0 timeout=120
> > primitive MySQL_Reader_VIP IPaddr2 \
> > params ip=10.0.1.66 cidr_netmask=32 nic=bond0.100 \
> > meta target-role=Started
> > primitive ping-gateway ocf:pacemaker:ping \
> > params host_list=10.0.1.1 multiplier=100 timeout=1 \
> > op monitor interval=10s timeout=20s
> > primitive resMON ocf:pacemaker:ClusterMon \
> > op start interval=0 timeout=90s \
> > op stop interval=0 timeout=100s \
> > op monitor interval=10s timeout=30s \
> > params extra_options="--mail-prefix MainDB_Cluster_Notification --mail-from cluster-alarm at gmsu.ua --mail-to cluster-alarm at gmsu.ua --mail-host mx.gmsu.ua"
> > ms MySQL_MasterSlave MySQL \
> > meta master-max=1 master-node-max=1 clone-max=2 clone-node-max=1 notify=true globally-unique=false target-role=Started is-managed=true
> > clone pingclone ping-gateway \
> > meta target-role=Started
> > location No-MySQL_Reader_VIP MySQL_Reader_VIP \
> > rule $id="No-MySQL_Reader_VIP-rule" -inf: readerOK eq 0 or not_defined readerOK
> > location cli-prefer-ClusterIP ClusterIP \
> > rule $id="cli-prefer-rule-ClusterIP" inf: #uname eq wb-db1
> > location cli-standby-MySQL_MasterSlave MySQL_MasterSlave \
> > rule $id="cli-standby-rule-MySQL_MasterSlave" -inf: #uname eq wb-db1
> > location resourceClusterIPwithping ClusterIP \
> > rule $id="resourceClusterIPwithping-rule" -inf: not_defined pingd or pingd lte 0
> > colocation MySQL_Reader_VIP_dislike_ClusterIP -200: MySQL_Reader_VIP ClusterIP
> > colocation MysqlMaster-with-ClusterIP inf: MySQL_MasterSlave:Master ClusterIP
> > order MysqlMaster-after-ClusterIP inf: ClusterIP MySQL_MasterSlave:promote
> > property cib-bootstrap-options: \
> > dc-version=1.1.10-14.el6_5.3-368c726 \
> > cluster-infrastructure="classic openais (with plugin)" \
> > expected-quorum-votes=2 \
> > no-quorum-policy=ignore \
> > stonith-enabled=false \
> > last-lrm-refresh=1402318675
> > property mysql_replication: \
> > MySQL_REPL_INFO="wb-db2|mysql-bin.000126|107"
> > rsc_defaults rsc-options: \
> > resource-stickiness=200
> >
> > Thank you!
> >
> >
> > 2014-06-05 3:17 GMT+03:00 Andrew Beekhof <andrew at beekhof.net>:
> >
> > On 30 May 2014, at 6:32 pm, Виталий Туровец <corebug at corebug.net> wrote:
> >
> > > Hello there, people!
> > > I am new to this list, so please excuse me if i'm posting to the wrong place.
> > >
> > > I've got a pacemaker cluster with such a configuration: http://pastebin.com/1SbWWh4n.
> > >
> > > Output of "crm status":
> > > ============
> > > Last updated: Fri May 30 11:22:59 2014
> > > Last change: Thu May 29 03:22:38 2014 via crmd on wb-db2
> > > Stack: openais
> > > Current DC: wb-db2 - partition with quorum
> > > Version: 1.1.7-6.el6-148fccfd5985c5590cc601123c6c16e966b85d14
> > > 2 Nodes configured, 2 expected votes
> > > 7 Resources configured.
> > > ============
> > >
> > > Online: [ wb-db2 wb-db1 ]
> > >
> > > ClusterIP (ocf::heartbeat:IPaddr2): Started wb-db2
> > > MySQL_Reader_VIP (ocf::heartbeat:IPaddr2): Started wb-db2
> > > resMON (ocf::pacemaker:ClusterMon): Started wb-db2
> > > Master/Slave Set: MySQL_MasterSlave [MySQL]
> > > Masters: [ wb-db2 ]
> > > Stopped: [ MySQL:1 ]
> > > Clone Set: pingclone [ping-gateway]
> > > Started: [ wb-db1 wb-db2 ]
> > >
> > > There was an unclean shutdown of a cluster and after that i've faced a problem that a slave of MySQL_MasterSlave resource does not come up.
> > > When i try to do a "cleanup MySQL_MasterSlave" i see such thing in logs:
> >
> > Most of those errors are cosmetic and fixed in later versions.
> >
> > > Version: 1.1.7-6.el6-148fccfd5985c5590cc601123c6c16e966b85d14
> >
> > It you can get to rhel 6.5 you'll have access to 1.1.10 where these are fixed.
> >
> > >
> > > May 29 03:22:22 [4423] wb-db1 crmd: warning: decode_transition_key: Bad UUID (crm-resource-4819) in sscanf result (3) for 0:0:crm-resource-4819
> > > May 29 03:22:22 [4423] wb-db1 crmd: warning: decode_transition_key: Bad UUID (crm-resource-4819) in sscanf result (3) for 0:0:crm-resource-4819
> > > May 29 03:22:22 [4423] wb-db1 crmd: info: ais_dispatch_message: Membership 408: quorum retained
> > > May 29 03:22:22 [4418] wb-db1 cib: info: set_crm_log_level: New log level: 3 0
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_ais_dispatch: Update relayed from wb-db2
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_ais_dispatch: Update relayed from wb-db2
> > > May 29 03:22:38 [4418] wb-db1 cib: info: apply_xml_diff: Digest mis-match: expected 2f5bc3d7f673df3cf37f774211976d69, calculated b8a7adf0e34966242551556aab605286
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_process_diff: Diff 0.243.4 -> 0.243.5 not applied to 0.243.4: Failed application of an update diff
> > > May 29 03:22:38 [4418] wb-db1 cib: info: cib_server_process_diff: Requesting re-sync from peer
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_server_process_diff: Not applying diff 0.243.4 -> 0.243.5 (sync in progress)
> > > May 29 03:22:38 [4418] wb-db1 cib: info: cib_replace_notify: Replaced: -1.-1.-1 -> 0.243.5 from wb-db2
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_trigger_update: Sending flush op to all hosts for: pingd (100)
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_trigger_update: Sending flush op to all hosts for: probe_complete (true)
> > > May 29 03:22:38 [4418] wb-db1 cib: info: set_crm_log_level: New log level: 3 0
> > > May 29 03:22:38 [4418] wb-db1 cib: info: apply_xml_diff: Digest mis-match: expected 754ed3b1d999e34d93e0835b310fd98a, calculated c322686deb255936ab54e064c696b6b8
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_process_diff: Diff 0.244.5 -> 0.244.6 not applied to 0.244.5: Failed application of an update diff
> > > May 29 03:22:38 [4418] wb-db1 cib: info: cib_server_process_diff: Requesting re-sync from peer
> > > May 29 03:22:38 [4423] wb-db1 crmd: info: delete_resource: Removing resource MySQL:0 for 4996_crm_resource (internal) on wb-db2
> > > May 29 03:22:38 [4423] wb-db1 crmd: info: notify_deleted: Notifying 4996_crm_resource on wb-db2 that MySQL:0 was deleted
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_server_process_diff: Not applying diff 0.244.5 -> 0.244.6 (sync in progress)
> > > May 29 03:22:38 [4423] wb-db1 crmd: warning: decode_transition_key: Bad UUID (crm-resource-4996) in sscanf result (3) for 0:0:crm-resource-4996
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_server_process_diff: Not applying diff 0.244.6 -> 0.244.7 (sync in progress)
> > > May 29 03:22:38 [4418] wb-db1 cib: notice: cib_server_process_diff: Not applying diff 0.244.7 -> 0.244.8 (sync in progress)
> > > May 29 03:22:38 [4418] wb-db1 cib: info: cib_replace_notify: Replaced: -1.-1.-1 -> 0.244.8 from wb-db2
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_trigger_update: Sending flush op to all hosts for: pingd (100)
> > > May 29 03:22:38 [4421] wb-db1 attrd: notice: attrd_trigger_update: Sending flush op to all hosts for: probe_complete (true)
> > > May 29 03:22:38 [4423] wb-db1 crmd: notice: do_lrm_invoke: Not creating resource for a delete event: (null)
> > > May 29 03:22:38 [4423] wb-db1 crmd: info: notify_deleted: Notifying 4996_crm_resource on wb-db2 that MySQL:1 was deleted
> > > May 29 03:22:38 [4423] wb-db1 crmd: warning: decode_transition_key: Bad UUID (crm-resource-4996) in sscanf result (3) for 0:0:crm-resource-4996
> > > May 29 03:22:38 [4423] wb-db1 crmd: warning: decode_transition_key: Bad UUID (crm-resource-4996) in sscanf result (3) for 0:0:crm-resource-4996
> > > May 29 03:22:38 [4418] wb-db1 cib: info: set_crm_log_level: New log level: 3 0
> > > May 29 03:22:38 [4423] wb-db1 crmd: info: ais_dispatch_message: Membership 408: quorum retained
> > >
> > > Here's the cibadmin -Q output from node that is alive: http://pastebin.com/aeqfTaCe
> > > And here's the one from failed node: http://pastebin.com/ME2U5vjK
> > > The question is: how do i somehow cleanup the things for master/slave resource MySQL_MasterSlave to start working properly?
> > >
> > > Thank you!
> > >
> > > --
> > >
> > >
> > >
> > >
> > > ~~~
> > > WBR,
> > > Vitaliy Turovets
> > > Lead Operations Engineer
> > > Global Message Services Ukraine
> > > +38(093)265-70-55
> > > VITU-RIPE
> > >
> > > _______________________________________________
> > > 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
> >
> >
> >
> >
> > --
> >
> >
> >
> >
> > ~~~
> > WBR,
> > Vitaliy Turovets
> > Lead Operations Engineer
> > Global Message Services Ukraine
> > +38(093)265-70-55
> > VITU-RIPE
> >
> > _______________________________________________
> > 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
>
>
>
>
> --
>
>
>
>
> ~~~
> WBR,
> Vitaliy Turovets
> Lead Operations Engineer
> Global Message Services Ukraine
> +38(093)265-70-55
> VITU-RIPE
>
> _______________________________________________
> 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
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 841 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20140627/ace4288b/attachment-0003.sig>
More information about the Pacemaker
mailing list