[Pgpool-general] difference between master and primary node in streaming mode

[Tatsuo Ishii]

> What's the difference between the master and primary node in streaming
> replication mode?
> 
> Occasionally, in my testing, I see this:
> 
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: find_primary_node_repeatedly:
> waiting for finding a primary node
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: last message repeated 10 times
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: find_primary_node: 0 node is standby
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: last message repeated 10 times
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: find_primary_node: primary node
> id is 1
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: failover: set new primary node: 1
> Oct  5 11:36:08 31214-1-807393 pgpool[28871]: failover: set new master node: 0
> 
> Note that the primary and master are different.  I understand that the primary
> node is considered whichever node is not in recovery mode, but what is the master?

The master node is the node which is alive and has the least node
id. Not important in streaming replication mode(important in pgpool-II
native replication mode, though).
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp