Horton

Re: locking [user] catalog tables vs 2pc vs logical rep

Amit Kapila <amit.kapila16@gmail.com>

View thread View on postgr.es
From: Amit Kapila <amit.kapila16@gmail.com>
To: Andres Freund <andres@anarazel.de>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>, Markus Wanner <markus.wanner@enterprisedb.com>
Date: 2021-02-23T03:54:18Z
Lists: pgsql-hackers 
On Tue, Feb 23, 2021 at 9:09 AM Andres Freund <andres@anarazel.de> wrote:
>
> On 2021-02-23 08:56:39 +0530, Amit Kapila wrote:
> > On Tue, Feb 23, 2021 at 3:58 AM Andres Freund <andres@anarazel.de> wrote:
> > > Perhaps all that we need to do is to disallow 2PC prepare if [user]
> > > catalog tables have been locked exclusively?
>
> > Right, and we have discussed this during development [1][2].
>
> I remember bringing it up before as well... Issues like this really need
> to be mentioned as explicit caveats at least somewhere in the code and
> commit message.
>

Okay, so is it sufficient to add comments in code, or do we want to
add something in docs? I am not completely sure if we need to add in
docs till we have core-implementation of prepare waiting to get
logically replicated.

-- 
With Regards,
Amit Kapila.

Commits

  1. Doc: Update caveats in synchronous logical replication.

  2. Document a few caveats in synchronous logical replication.

  3. Update the docs and comments for decoding of prepared xacts.

  4. Allow decoding at prepare time in ReorderBuffer.

  5. Documentation for logical decoding.