Horton

Re: Use of backup_label not noted in log

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: David Steele <david@pgmasters.net>
Cc: Michael Banck <mbanck@gmx.net>, Michael Paquier <michael@paquier.xyz>, Andres Freund <andres@anarazel.de>, Pg Hackers <pgsql-hackers@postgresql.org>, Laurenz Albe <laurenz.albe@cybertec.at>
Date: 2024-01-25T21:42:17Z
Lists: pgsql-hackers 
David Steele <david@pgmasters.net> writes:
> Another thing to note here -- knowing the LSN is important but also 
> knowing that backup recovery was attempted (i.e. backup_label exists) is 
> really crucial. Knowing both just saves so much time in back and forth 
> debugging.

> It appears the tally for back patching is:

> For: Andres, David, Michael B
> Not Sure: Robert, Laurenz, Michael P

> It seems at least nobody is dead set against it.

We're talking about 1d35f705e, right?  That certainly looks harmless
and potentially useful.  I'm +1 for back-patching.

			regards, tom lane

Commits

  1. Add more LOG messages when starting and ending recovery from a backup