Re: Can we change pg_rewind used without wal_log_hints and data_checksums

From: Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>
To: lchch1990(at)sina(dot)cn, pgsql-hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Can we change pg_rewind used without wal_log_hints and data_checksums
Date: 2026-01-15 19:08:25
Message-ID: b9ac506be4222972140684c7cd549fa1c7a2e341.camel@cybertec.at
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Thu, 2026-01-15 at 14:14 +0800, lchch1990(at)sina(dot)cn wrote:
> > See https://postgr.es/m/CA%2BTgmoY4j%2Bp7JY69ry8GpOSMMdZNYqU6dtiONPrcxaVG%2BSPByg%40mail.gmail.com
>
> Yes I known the step and I have check the mail link. 
> As described in the top mail we can find some way to solve the problem so that 
> pg_rewind can run without wal_log_hints and data_checksums. 
>
> Currently pg_rewind search wal start at checkpoint lsn or redo lsn, I mean to search more 
> wal to cover whole releated transactions so any releated pages with copyed, and we never 
> warried about hint bits issue. 

I apologize for my misunderstanding.

I had a brief look at the patch, and the gratuitous use of static variables didn't
appeal to me. Can you briefly describe the algorithm? You look at all commit
records *after* the fork, right? Then how can you identify how far back you have
rewind? How can you identify when a transaction started?

Yours,
Laurenz Albe

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Dave Cramer 2026-01-15 19:11:22 Re: Proposal to allow setting cursor options on Portals
Previous Message Nathan Bossart 2026-01-15 19:07:18 Re: refactor architecture-specific popcount code