Horton

Re: pendingOps table is not cleared with fsync=off

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Heikki Linnakangas <hlinnaka@iki.fi>
Cc: Thomas Munro <thomas.munro@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2020-08-06T18:10:15Z
Lists: pgsql-hackers 
Heikki Linnakangas <hlinnaka@iki.fi> writes:
> On 06/08/2020 18:42, Tom Lane wrote:
>> I'm looking at this commit in connection with writing release notes
>> for next week's releases.  Am I right in thinking that this bug leads
>> to indefinite bloat of the pendingOps hash table when fsync is off?
>> If so, that seems much more worth documenting than the assertion
>> failure.

> That's correct.

OK, thanks for confirming.

			regards, tom lane

Commits

  1. Move check for fsync=off so that pendingOps still gets cleared.

  2. Refactor the fsync queue for wider use.