RE: Synchronizing slots from primary to standby

From: "Hayato Kuroda (Fujitsu)" <kuroda(dot)hayato(at)fujitsu(dot)com>
To: 'Amit Kapila' <amit(dot)kapila16(at)gmail(dot)com>
Cc: shveta malik <shveta(dot)malik(at)gmail(dot)com>, "Zhijie Hou (Fujitsu)" <houzj(dot)fnst(at)fujitsu(dot)com>, Peter Smith <smithpb2250(at)gmail(dot)com>, "Drouvot, Bertrand" <bertranddrouvot(dot)pg(at)gmail(dot)com>, Nisha Moond <nisha(dot)moond412(at)gmail(dot)com>, Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com>, Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>, Bruce Momjian <bruce(at)momjian(dot)us>, Ashutosh Sharma <ashu(dot)coek88(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Ajin Cherian <itsajin(at)gmail(dot)com>, Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>
Subject: RE: Synchronizing slots from primary to standby
Date: 2023-12-26 09:30:47
Message-ID: TY3PR01MB9889112F25F32C89FCFF3368F598A@TY3PR01MB9889.jpnprd01.prod.outlook.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Dear Amit,

> I think we should be able to detect it if we want but do we want to
> add this restriction considering that users can always install the
> required plugins after standby gets promoted? I think we can do either
> way in this case but as we are not going to use these slots till the
> standby node is promoted, it seems okay to validate the plugins after
> promotion once users use the synced slots.

Personally it should be detected, but I want to hear opinions from others.
Below are my reasons:

1)
We can avoid a possibility that users miss the installation of plugins. Basically
we should detect before the issue will really occur.

2)
Rules around here might be inconsistent. Slots which will be synchronized can be
created either way:

a) manual creation via SQL function, or
b) automatic creation by slotsync worker.

In case of a), the decoding context is created when creation so that the plugin
must be installed. Case b), however, we allow not to install beforehand. I felt
it might be confused for users. Thought?

Best Regards,
Hayato Kuroda
FUJITSU LIMITED

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrei Lepikhov 2023-12-26 09:37:51 Specify description of the SpecialJoinInfo structure
Previous Message ywgrit 2023-12-26 09:25:05 Re: planner chooses incremental but not the best one