[pgpool-general: 1053] consecutive failovers do not work

Amador Alvarez aalvarez at d2.com
Sat Sep 29 02:36:56 JST 2012


Hi there,

I have two servers running on postgres 9 streaming replication, 
wondering yet either 2 pgpools or one though, supposing 1 pgpool running 
over pacemaker for the high availability I am wondering how to deal with 
consecutive failovers as the failover script creates a trigger file when 
crashes the node 0, however after the failover the node 1 becomes the 
master and if another failover comes up the script would not generate a 
trigger file as it is not the node 0 anymore.
Is this thought right?

I would appreciate any insight on this matter,

Thanks in advance,
A.A.


More information about the pgpool-general mailing list