Feedback-ID: id4a34324:Fastmail
Date: Mon, 24 Apr 2023 17:16:23 -0700
From: Andres Freund <andres@anarazel.de>
To: Christoph Berg <myon@debian.org>, Thomas Munro <thomas.munro@gmail.com>
Cc: pgsql-hackers@lists.postgresql.org
Subject: Re: could not extend file "base/5/3501" with FileFallocate():
 Interrupted system call
Message-ID: <20230425001623.im5uquriqlk4ayy6@awork3.anarazel.de>
References: <E1pk6ud-001WhT-N8@gemulon.postgresql.org>
 <ZEZDj1H61ryrmY9o@msg.df7cb.de>
 <20230424223225.gle3x3sw5m4div2l@awork3.anarazel.de>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="hfuy3uja5qh36y7g"
Content-Disposition: inline
In-Reply-To: <20230424223225.gle3x3sw5m4div2l@awork3.anarazel.de>
Archived-At: 
 <https://www.postgresql.org/message-id/20230425001623.im5uquriqlk4ayy6%40awork3.anarazel.de>
Precedence: bulk


--hfuy3uja5qh36y7g
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

Hi,

On 2023-04-24 15:32:25 -0700, Andres Freund wrote:
> On 2023-04-24 10:53:35 +0200, Christoph Berg wrote:
> > I'm often seeing PG16 builds erroring out in the pgbench tests:
> > I don't think the disk is full since it's always hitting that same
> > spot, on some of the builds:
> 
> Yea, the EINTR pretty clearly indicates that it's not really out-of-space.

FWIW, I tried to reproduce this, without success - not too surprising, I
assume it's rather timing dependent.


> We obviously can add a retry loop to FileFallocate(), similar to what's
> already present e.g. in FileRead(). But I wonder if we shouldn't go a bit
> further, and do it for all the fd.c routines where it's remotely plausible
> EINTR could be returned? It's a bit silly to add EINTR retries one-by-one to
> the functions.
> 
> 
> The following are documented to potentially return EINTR, without fd.c having
> code to retry:
> 
> - FileWriteback() / pg_flush_data()
> - FileSync() / pg_fsync()
> - FileFallocate()
> - FileTruncate()
> 
> With the first two there's the added complication that it's not entirely
> obvious whether it'd be better to handle this in File* or pg_*. I'd argue the
> latter is a bit more sensible?

A prototype of that approach is attached. I pushed the retry handling into the
pg_* routines where applicable.  I guess we could add pg_* routines for
FileFallocate(), FilePrewarm() etc as well, but I didn't do that here.

Christoph, could you verify this fixes your issue?

Greetings,

Andres Freund

--hfuy3uja5qh36y7g
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v1-0001-fd.c-Retry-after-EINTR-in-more-places.patch"