[ClusterLabs] Pacemaker with PostgreSQL 9.6

[Barrow Kwan]

Hi,

                I have been using Pacemaker + PostgreSQL 9.4 for many year without any issue.  Recently, I setup another cluster with Pacemaker + PostgreSQL 9.6 on CentOS 6.  However the cluster didn’t seem to have problem to set the Slave’s score from “-INFINITY” to “100”.  When the master is down, slave can’t be promoted to Master.  Did anyone have similar issue?  Here is what the crm_mon -Afr show.

 

 

 

Stack: cman

Current DC: postgres-n01 (version 1.1.18-3.el6-bfe4e80420) - partition with quorum

Last updated: Mon Oct  8 14:38:49 2018

Last change: Mon Oct  8 14:38:45 2018 by root via crm_attribute               on postgres-n01

 

2 nodes configured

4 resources configured

 

Online: [ postgres-n01 postgres-n02 ]

 

Full list of resources:

 

 Master/Slave Set: pgsqld-master [pgsqld]

     Masters: [ postgres-n01 ]

     Slaves: [ postgres-n02 ]

 Resource Group: master-group

     pgsql-master-ip    (ocf::heartbeat:IPaddr2):     Started postgres-n01

 

Node Attributes:

* Node postgres-n01:

    + master-pgsqld                     : 1000

    + pgsqld-data-status                : LATEST

    + pgsqld-master-baseline            : 0000000004000300

    + pgsqld-status                     : PRI

* Node postgres-n02:

    + master-pgsqld                     : -INFINITY

    + pgsqld-data-status                : STREAMING|ASYNC

    + pgsqld-status                     : HS:async

 

Migration Summary:

* Node postgres-n02:

* Node postgres-n01:

 

 

 

-- 

Barrow Kwan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20181008/fe3bf07a/attachment.html>