Horton

Re: Reorderbuffer crash during recovery

Alvaro Herrera <alvherre@2ndquadrant.com>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: vignesh C <vignesh21@gmail.com>
Cc: Dilip Kumar <dilipbalaut@gmail.com>, Amit Kapila <amit.kapila16@gmail.com>, Andres Freund <andres@anarazel.de>, Tomas Vondra <tomas.vondra@2ndquadrant.com>, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2020-01-17T21:12:16Z
Lists: pgsql-bugs, pgsql-hackers 
On 2020-Jan-17, vignesh C wrote:

> Thanks Dilip for reviewing.
> I have fixed the comments you have suggested.

I ended up rewording that comment completely; I thought the original was
not explaining things well enough.

I also changed the comment for final_lsn in reorderbuffer.h: not only I
remove the line that I added in df9f682c7bf8 (the previous bugfix), but
I also remove the line that says "error during decoding", which came in
with the very first logical decoding commit (b89e151054a); I couldn't
find any evidence that final_lsn is being set on errors of any kind
(other than transaction abort, which doesn't seem an "error" in that
sense.)

Please give these comments a read; maybe I have misunderstood something
and my comment is wrong.

Pushed now to all branches.  Thanks, Vignesh, Amit and Dilip.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Commits

  1. Set ReorderBufferTXN->final_lsn more eagerly

  2. Fix failure to delete spill files of aborted transactions