[pgpool-general: 8849] Re: pgpool_remote_start won't start pgpool at all

[Bo Peng]

Hi,

Sorry for the late reply.

> I am confused about the procedures to recover a node.
> 
> https://www.pgpool.net/docs/latest/en/html/runtime-online-recovery.html
> 
> It is triggered by manually sending `pcp_recovery` on the  current master
> server to the current master pgpool. I totally understand that the first
> stage is to evoke the postgres replication process to copy data from the
> current master to the new replica.
> 
> But here are 2 questions:
> 
> 1) Since the 2nd step requires there is no real client (aside from other
> pgpool's  healthcheck ones) , is it possible to start the 2nd stage?
> especially for snapshot mode
> 
> *Note: * There is a restriction in the online recovery in native
> replication mode
> <https://www.pgpool.net/docs/latest/en/html/runtime-config-running-mode.html#GUC-REPLICATION-MODE>.
> If Pgpool-II itself is installed on multiple hosts, online recovery does
> not work correctly, because Pgpool-II has to stop all the clients during
> the 2nd stage of online recovery. If there are several Pgpool-II hosts,
> only one of them will have received the online recovery command and will
> block the connections from clients.

If you are using multiple pgpool with watchdog enabled,
you can ignore the above description.
Watchdog will handle that.
 
> 2) After the 2nd stage, pgpool goes to pgpool_remote_start scripts to
> finish the recovery procedures. But pgpool_remote_start only starts the
> postgres on the replica server, while it does not do anything to the pgpool
> on that server at all.
> 
> How to start the pgpool server? Just start it or take some registration
> procedure to let the master pgpool welcome it into the whole pgpools'
> cluster?

pcp_recovery_node is used to recover and start
a failed PostgreSQL standby, not pgpool.
You have to manually start all pgpool on each node.

-- 
Bo Peng <pengbo at sraoss.co.jp>
SRA OSS LLC
TEL: 03-5979-2701 FAX: 03-5979-2702
URL: https://www.sraoss.co.jp/