Re: A micro-optimisation for walkdir() - Mailing list pgsql-hackers

From Thomas Munro
Subject Re: A micro-optimisation for walkdir()
Date
Msg-id CA+hUKGJD4c+45o8vcvi073ocd-Y5As8f2MhYz-ndzdXGxqiRKA@mail.gmail.com
Whole thread Raw
In response to Re: A micro-optimisation for walkdir()  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: A micro-optimisation for walkdir()  (Juan José Santamaría Flecha <juanjo.santamaria@gmail.com>)
List pgsql-hackers
On Tue, Sep 8, 2020 at 10:53 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Thomas Munro <thomas.munro@gmail.com> writes:
> > FWIW I just spotted a couple of very suspicious looking failures for
> > build farm animal "walleye", a "MinGW64 8.1.0" system, that say:
>
> walleye's been kind of unstable since the get-go, so I wouldn't put
> too much faith in reports just from it.

CC'ing animal owner.

<pokes at the logs>  I suspect that someone who knows about PostgreSQL
on Windows would recognise the above symptom, but my guess is the
Windows "indexing" service is on, or an antivirus thing, or some other
kind of automatically-open-and-sniff-every-file-on-certain-file-events
thing.  It looks like nothing of ours is even running at that moment
("waiting for server to shut down.... done"), and it's the RMDIR /s
shell command that is reporting the error.  The other low probability
error seen on this host is this one:

+ERROR:  could not stat file "pg_wal/000000010000000000000007":
Permission denied



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: PATCH: logical_work_mem and logical streaming of large in-progress transactions
Next
From: Tom Lane
Date:
Subject: Re: Gripes about walsender command processing