Thread: Mutex error 22 - Postgres version 14

sireesha <sireesha.padmini@gmail.com> writes:
> We recently upgraded our production databases to 14 and we have encountered
> Mutex 22 error and Postgres defunct processes in 2 of the databases.

There is no part of Postgres that would produce a message like "Mutex 22
error".  You need to spend a bit more effort on identifying where your
problem is coming from.  If it does seem to be coming from Postgres,
you need to spend a lot more effort on providing a useful trouble report
if you want any help from the mailing lists.  Please see

https://wiki.postgresql.org/wiki/Guide_to_reporting_problems

            regards, tom lane

sireesha <sireesha.padmini@gmail.com> writes:
> This error is reported in Postgresql log and exact message from the log is
> below .

> 2023-01-24 02:35:45.833 PST [3424807] LOG:  PID 0 in cancel request did not
> match any process
> *Error locking mutex 22*

The first of those lines comes from this bit in postmaster.c:

    /* No matching backend */
    ereport(LOG,
            (errmsg("PID %d in cancel request did not match any process",
                    backendPID)));

As you can see, that would not have generated anything about a mutex.
The string "locking mutex" appears nowhere in the Postgres sources;
in fact, so far as I can find we don't use the word "mutex" in any
message whatever.  So that second line is coming from something else.
Given that it's showing up in postmaster stderr, it might be coming
from libc, or from some third-party extension.  But with zero context
about your system, it's hard for anyone to guess what exactly.

> Please let me know if i have to provide any other trouble report.

I take it you still didn't read the "Guide to reporting problems".
You need to err on the side of providing more information, not less.

            regards, tom lane

On Wed, Feb 1, 2023 at 3:02 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> > 2023-01-24 02:35:45.833 PST [3424807] LOG:  PID 0 in cancel request did not
> > match any process
> > *Error locking mutex 22*
>
> The first of those lines comes from this bit in postmaster.c:
>
>     /* No matching backend */
>     ereport(LOG,
>             (errmsg("PID %d in cancel request did not match any process",
>                     backendPID)));
>
> As you can see, that would not have generated anything about a mutex.
> The string "locking mutex" appears nowhere in the Postgres sources;
> in fact, so far as I can find we don't use the word "mutex" in any
> message whatever.

I wonder if 22 might be EINVAL, which is one possible error code used
by pthread_mutex_lock().

-- 
Peter Geoghegan

Peter Geoghegan <pg@bowt.ie> writes:
> On Wed, Feb 1, 2023 at 3:02 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> 2023-01-24 02:35:45.833 PST [3424807] LOG:  PID 0 in cancel request did not
>>> match any process
>>> *Error locking mutex 22*

> I wonder if 22 might be EINVAL, which is one possible error code used
> by pthread_mutex_lock().

Maybe, but we still don't know what's reporting the error.

I tried searching for "Error locking mutex" in Debian Code Search,
and got several hits, but none of them match this exactly --- the
ones that offer any additional info present it as a string not a
number.

            regards, tom lane

sireesha <sireesha.padmini@gmail.com> writes:
> Here is the detailed information of the error.
> There is an additional error *"Error WriteLocking RWLock!35" along with
> Mutex 22 error in the logfile.*

Well, that's *another* string that certainly did not come out of Postgres
... and if debian code search is to be trusted, it didn't come out of
repmgr or pgbouncer either.  I speculate you have some other extension(s)
installed that you've not told us about.

            regards, tom lane