Horton

Re: Avoid erroring out when unable to remove or parse logical rewrite files to save checkpoint work

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: "Bossart, Nathan" <bossartn@amazon.com>
Cc: Andres Freund <andres@anarazel.de>, Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>, Julien Rouhaud <rjuju123@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-01-19T22:53:43Z
Lists: pgsql-hackers 
"Bossart, Nathan" <bossartn@amazon.com> writes:
> I think the other side of this is that we don't want checkpointing to
> continually fail because of a noncritical failure.  That could also
> lead to problems down the road.

Yeah, a persistent failure to complete checkpoints is very nasty.
Your disk will soon fill with unrecyclable WAL.  I don't see how
that's better than a somewhat hypothetical performance issue.

			regards, tom lane

Commits

  1. Expand the use of get_dirent_type(), shaving a few calls to stat()/lstat()

  2. fsync pg_logical/mappings in CheckPointLogicalRewriteHeap().

  3. Introduce logical decoding.