Обсуждение: server process (PID ) was terminated by signal 29

Hi all,

After upgrading one of our database servers from 8.1.11 on SLES9 to
8.2.6 on CentOS 5.1, it crashed several times with following errors:

LOG:  server process (PID ) was terminated by signal 29
LOG:  terminating any other active server processes
WARNING:  terminating connection because of crash of another server process
DETAIL:  The postmaster has commanded this server process to roll back
the current transaction and exit, because another server process exited
abnormally and possibly corrupted shared memory.
HINT:  In a moment you should be able to reconnect to the database and
repeat your command.
........

LOG:  all server processes terminated; reinitializing

Does this mean bad hardware, corrupted database or something else?

TIA,
Milen

Milen Evtimov <milen@netinfo.bg> writes:
> After upgrading one of our database servers from 8.1.11 on SLES9 to
> 8.2.6 on CentOS 5.1, it crashed several times with following errors:

> LOG:  server process (PID ) was terminated by signal 29

That's pretty odd.  Would you confirm what signal 29 is on your machine?
(On my Fedora 8 box, the numbers are defined in /usr/include/bits/signum.h
--- it's probably the same on yours.)

Fedora 8 says that 29 is SIGIO a/k/a SIGPOLL, but that doesn't make any
sense to me --- we don't use any calls that would trigger that.

Is this a locally built copy of Postgres, or an RPM installation?
If the latter, which RPM exactly?

            regards, tom lane

Tom Lane wrote:

> That's pretty odd.  Would you confirm what signal 29 is on your machine?
> (On my Fedora 8 box, the numbers are defined in /usr/include/bits/signum.h
> --- it's probably the same on yours.)

FWIW the easiest way to get the signal names is kill -l:

$ kill -l 29
IO

--
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

Tom Lane wrote:
> Milen Evtimov <milen@netinfo.bg> writes:
>
>> After upgrading one of our database servers from 8.1.11 on SLES9 to
>> 8.2.6 on CentOS 5.1, it crashed several times with following errors:
>>
>
>
>> LOG:  server process (PID ) was terminated by signal 29
>>
>
> That's pretty odd.  Would you confirm what signal 29 is on your machine?
> (On my Fedora 8 box, the numbers are defined in /usr/include/bits/signum.h
> --- it's probably the same on yours.)
>
> Fedora 8 says that 29 is SIGIO a/k/a SIGPOLL, but that doesn't make any
> sense to me --- we don't use any calls that would trigger that.
>
> Is this a locally built copy of Postgres, or an RPM installation?
> If the latter, which RPM exactly?
>
>             regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
>        choose an index scan if your joining column's datatypes do not
>        match
>
>
>
 From /usr/include/bits/signum.h
#define SIGPOLL         SIGIO   /* Pollable event occurred (System V).  */
#define SIGIO           29      /* I/O now possible (4.2 BSD).  */

and output from kill -l 29 is IO

Postgresql is installed from RPMs taken from official postgresql.org
site - exact version is
postgresql-8.2.6-1PGDG.rhel5.x86_64.

Regards,
Milen