Re: postmaster became multithreaded

From: Noah Misch <noah(at)leadboat(dot)com>
To: Wayne Pichotta <wpichotta(at)truecar(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: "pgsql-bugs(at)postgresql(dot)org" <pgsql-bugs(at)postgresql(dot)org>
Subject: Re: postmaster became multithreaded
Date: 2017-04-12 02:56:00
Message-ID: 20170412025600.GA2891953@tornado.leadboat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Tue, Mar 28, 2017 at 04:32:44PM -0400, Tom Lane wrote:
> Wayne Pichotta <wpichotta(at)truecar(dot)com> writes:
> > I just saw this in my logs:
> > LOG: postmaster became multithreaded
> > DETAIL: Please report this to <pgsql-bugs(at)postgresql(dot)org>.
>
> Just reporting the bare fact of that message isn't helpful.

You can blame me for writing the message that way. Wayne, thanks for
following the message's direction.

> The known causes of that involve doing something thread-oriented
> in an "unsafe" language such as pltclu, plperlu, or plpythonu,
> or perhaps bare C --- and even then, you'd have had to go well out
> of your way to make it happen in the postmaster process
> instead of some child process.

/*
* On macOS, libintl replaces setlocale() with a version that calls
* CFLocaleCopyCurrent() when its second argument is "" and every relevant
* environment variable is unset or empty. CFLocaleCopyCurrent() makes
* the process multithreaded.

The reported message might be explained by starting the server with
lc_messages='something' in postgresql.conf, then deleting that line from
postgresql.conf and reloading the configuration. Wayne, does that sound like
something you might have done? If not ...

> However, you haven't told us what nondefault configuration you
> might be using, or what platform this is on, or even what PG version
> it is. So it's impossible to speculate further.

... please post the output of the "pg_config" command and describe anything
unusual about your configuration. (We do know this is macOS.)

> FWIW, that's only an informational log message emitted during a
> postmaster shutdown that occurred for some other reason. Maybe
> it's related to the reason, but again we have zero info. Preceding
> lines in the log might have provided some knowledge.

Agreed, a larger portion of log would help.

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Noah Misch 2017-04-12 04:32:24 Re: postmaster became multithreaded
Previous Message shao bret 2017-04-12 02:17:11 答复: [BUGS] 答复: [BUGS] BUG #14615: ReplicationOriginShmemInit Memory access cross-border