[pgpool-general: 6324] Re: Failover issue while manually promoting to Master !

[Bo Peng]

Hi,

On Fri, 30 Nov 2018 17:23:33 +0530
Nitish Kumar <itcell.mpwz at mp.gov.in> wrote:

> Hi All,
> 
> I am using PgPool II 3.7.5 with PostgreSQL 10.5 nodes in the backend.
> 
> My DB cluster is running in Master Slave Hot Standby mode with 1 master and 3 slaves.
> 
> My PgPool II is also running in Master Slave/Streaming mode. Today I have stopped services in Master PostgreSQL Node and manually promoted
> one of the slave node to be Master Node.
> 
> After that I have restarted PostgreSQL Nodes expect the original Master and PgPool II services.

During you restart PostgreSQL nodes, Pgpool-II couldn't detect backend nodes,
so Pgpool-II thinks all backend nodes are down and closed the connections.

Before you restart PostgreSQL nodes, you need stop Pgpool-II first.
 
> Everything is Up but when I try to connect to PgPool II from my REST API application this is the error I am getting:
> 
> 
> .postgresql.util.PSQLException: FATAL: failed to create a backend connection
>  Detail: executing failover on backend
> 
> 
> PgPool II services is Up and running. PostgreSQL nodes are up and running except the old master.
> 
> What is that I am doing wrong. Kindly help and guide.
> Regards,
> Nitish Kumar


-- 
Bo Peng <pengbo at sraoss.co.jp>
SRA OSS, Inc. Japan