Re: [SQL] Re: Very disappointing performance -- All Indexes Ignored.

From: secret <secret(at)kearneydev(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-sql(at)hub(dot)org
Subject: Re: [SQL] Re: Very disappointing performance -- All Indexes Ignored.
Date: 1999-03-12 18:55:33
Message-ID: 36E96325.F99A51FA@kearneydev.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql

Tom Lane wrote:

> secret <secret(at)kearneydev(dot)com> writes:
> > PostgreSQL is also crashing 1-2 times a day on me, although I have a
> > handy perl script to keep it alive now <grin>...
>
> Details please? 6.4.x has been pretty stable for me, and for most other
> people as far as I've heard. If you're getting a backend coredump,
> a stack backtrace with gdb or equivalent would be helpful info.
>
> regards, tom lane

I sent in a bug report on it. Would a -d x report be helpful? I included
a -d 3 but if a higher level would be better I'll surely include that,
basically the server randomly dies with a:

ERROR: postmaster: StreamConnection: accept: Invalid argument
pmdie 3
(then signals all children to drop dead)

And all the postmasters go ZOMBIE. It doesn't core as far as I can tell,
it just
dies at the invalid argument and brings everything crashing down behind it.
Could the ODBC driver be sending something goofy that makes it coke? It's
very irregular and I haven't been able to get an ODBC trace yet.

I did however find a fun new CREATE VIEW which crashes the backend 100%,
it's on the bugs list with a 'method of reproduction' that worked for me 100%,
I think it'll probably be an easy fix.(It doesn't zombie all the postmasters
like the other one tho)

--David

In response to

Browse pgsql-sql by date

  From Date Subject
Next Message Tim Perdue 1999-03-12 23:35:17 So what happens at 2GB?
Previous Message Tom Lane 1999-03-12 15:57:31 Re: Very disappointing performance -- All Indexes Ignored.