Re: Checkpoint not retrying failed fsync?

From: Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com>
To: Andrew Gierth <andrew(at)tao11(dot)riddles(dot)org(dot)uk>
Cc: Pg Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Checkpoint not retrying failed fsync?
Date: 2018-04-05 23:17:48
Message-ID: CAEepm=2ubZitOjJYBZq9fLu-zL_04YauC3+j35MqLNK3wR3-aQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox
Thread:
Lists: pgsql-hackers

On Fri, Apr 6, 2018 at 10:16 AM, Andrew Gierth
<andrew(at)tao11(dot)riddles(dot)org(dot)uk> wrote:
> Furthermore, checking the trace output from the checkpointer process, it
> is not even attempting an fsync of the failing file; this isn't like the
> Linux fsync issue, I've confirmed that fsync will repeatedly fail on the
> file until the underlying errors stop.

Thank you for confirming that! Now, how does one go about buying
shares in FreeBSD?

> As far as I can tell from reading the code, if a checkpoint fails the
> checkpointer is supposed to keep all the outstanding fsync requests for
> next time. Am I wrong, or is there some failure in the logic to do this?

Yikes. I think this is suspicious:

* The bitmap manipulations are slightly tricky,
because we can call
* AbsorbFsyncRequests() inside the loop and that
could result in
* bms_add_member() modifying and even re-palloc'ing
the bitmapsets.
* This is okay because we unlink each bitmapset from
the hashtable
* entry before scanning it. That means that any incoming fsync
* requests will be processed now if they reach the
table before we
* begin to scan their fork.

Why is it OK to unlink the bitmapset? We still need its contents, in
the case that the fsync fails!

--
Thomas Munro
http://www.enterprisedb.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2018-04-05 23:33:55 Re: [HACKERS] path toward faster partition pruning
Previous Message Claudio Freire 2018-04-05 22:59:59 Re: Vacuum: allow usage of more than 1GB of work mem