| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Simon Riggs <simon(at)2ndquadrant(dot)com>, Nikhil Sontakke <nikhils(at)2ndquadrant(dot)com>, Ajin Cherian <itsajin(at)gmail(dot)com> |
| Subject: | Re: Track replica origin progress for Rollback Prepared |
| Date: | 2021-01-05 07:02:07 |
| Message-ID: | X/QO71SVVDPcc7IL@paquier.xyz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, Jan 05, 2021 at 09:35:21AM +0530, Amit Kapila wrote:
> As noted in [1], without this the subscriber might again ask for
> rollback prepared lsn after restart.
>
> Attached patch addresses this problem.
Is it possible to add some tests in test_decoding?
/* dump transaction origin information only for abort prepared */
if ((replorigin_session_origin != InvalidRepOriginId) &&
- TransactionIdIsValid(twophase_xid) &&
- XLogLogicalInfoActive())
+ TransactionIdIsValid(twophase_xid))
It seems to me that you may want to document as a comment the reason
why this gets sent even if (wal_level < logical).
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Luc Vlaming | 2021-01-05 07:13:07 | Re: Parallel Inserts in CREATE TABLE AS |
| Previous Message | Michael Paquier | 2021-01-05 06:47:59 | Re: Moving other hex functions to /common |