[pgpool-general: 8842] Re: What criteria is for a `failover` and what for a `recovery`?

[Zhaoxun Yan]

Hi Bo!

I am still confused about what to trigger 'recovery' script:

Is it triggered on pgpool to detect that there is a postgresql server on a
previously  failed host running on the default port?
Or it will not wait until such a server resumes running, but initiate
recovery_1st again and again until finally it succeeds?
These are totally different.




On Thu, Jun 15, 2023 at 7:45 AM Bo Peng <pengbo at sraoss.co.jp> wrote:

> Hi,
>
> > Suppose I have configured ` health_check_user` , health_check_password,
> > health_check_database & health_check_max_retries
> >
> > Suppose now no postgres server is up. Will starting a pgpool
> automatically
> > detect the flaw and execute the failover script, even if it yields no
> > result?
>
> Yes. Failover will be performed and pgpool will not accept any requests.
> Before starting pgpool, we need to make sure one primary is available.
>
> > About `recovery` same thing again. Will the postgres server starts itself
> > before  pgpool recognize it as a `recovery node`? In this case, should
> > pgpool stop this new server before starting the replication?
> >
> > In general, how do you define the trigger event for each?
>
> Pgpool-II's "online recovery" is used to recovery a failed (DOWN) standby
> node.
> Pgpool-II's "online recovery" performs the recovery scripts which will
> automatically
> configure streaming replication.
>
> The following documents may help you configure pgpool.
>
> https://www.pgpool.net/docs/latest/en/html/example-cluster.html
> https://www.pgpool.net/docs/latest/en/html/example-replication-mode.html
> --
> Bo Peng <pengbo at sraoss.co.jp>
> SRA OSS LLC
> TEL: 03-5979-2701 FAX: 03-5979-2702
> URL: https://www.sraoss.co.jp/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-general/attachments/20230616/83d30a68/attachment.htm>