Horton

Re: potential stuck lock in SaveSlotToPath()

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
Cc: Robert Haas <robertmhaas@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>, Andres Freund <andres@anarazel.de>, Alvaro Herrera <alvherre@2ndquadrant.com>
Date: 2020-03-27T07:48:45Z
Lists: pgsql-hackers 
On Thu, Mar 26, 2020 at 02:16:05PM +0100, Peter Eisentraut wrote:
> committed and backpatched

The patch committed does that in three places:
    /* rename to permanent file, fsync file and directory */
    if (rename(tmppath, path) != 0)
    {
+       LWLockRelease(&slot->io_in_progress_lock);
	ereport(elevel,
                (errcode_for_file_access(),
                 errmsg("could not rename file \"%s\" to \"%s\": %m",

But why do you assume that LWLockRelease() never changes errno?  It
seems to me that you should save errno before calling LWLockRelease(),
and then restore it back before using %m in the log message, no?  See
for example the case where trace_lwlocks is set.
--
Michael

Commits

  1. Drop slot's LWLock before returning from SaveSlotToPath()