AW: FATAL ERROR

In Addition of the last Mail:

After Migrating to PostgreSQL 7.0.3
the Error still occurs :-(

Greetings,

Michael J. Roth


-----Ursprüngliche Nachricht-----
Von: Michael J. Roth [mailto:MichaelRoth@schmuecker.de]
Gesendet: Dienstag, 12. Juni 2001 20:33
An: 'Tom Lane'
Cc: 'pgsql-novice@postgresql.org'
Betreff: AW: [NOVICE] FATAL ERROR


Hi,

thanks for your answer :-)

Here are two logfiles, one in debuglevel 2, one
in debuglevel 5.
Both files are packed with gzip.

We have a look to the logfiles before sending now,
but we only see, that the processes exited with 0.
Dont know, why ...

Best Regards,
Michael J. Roth


-----Ursprüngliche Nachricht-----
Von: pgsql-novice-owner@postgresql.org
[mailto:pgsql-novice-owner@postgresql.org]Im Auftrag von Tom Lane
Gesendet: Dienstag, 12. Juni 2001 08:28
An: MichaelRoth@schmuecker.de
Cc: pgsql-novice@postgresql.org
Betreff: Re: [NOVICE] FATAL ERROR


"Michael Roth" <MichaelRoth@schmuecker.de> writes:
> Sometimes we got the following error-message:
> PostgresSQL query failed: FATAL 1: Socket command type 2 unknown

Curious.  This seems to indicate that the backend has gotten out of sync
with the client, and is trying to interpret some data out of the middle
of a message as the start of a fresh command message.  We used to have
some bugs that would let that happen occasionally, but I thought they'd
all been fixed as of 7.0.*.  It would be useful to try to track down
what happens just before this message appears.  Could you run with query
logging enabled (start postmaster with -d2, and make sure its stderr
output goes somewhere useful)?

> PSQL-Version: 7.0.2

I'd advise upgrading to 7.0.3 (or even 7.1.2, but 7.0.3 would be a
painless update).  There are some nasty bugs fixed as of 7.0.3.
I don't think any of them explain this problem ... but I might be wrong.

            regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?

http://www.postgresql.org/search.mpl