Core dump on HP

From: Michael Brusser <michael(at)synchronicity(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Core dump on HP
Date: 2003-06-23 20:45:10
Message-ID: DEEIJKLFNJGBEMBLBAHCAELEDEAA.michael@synchronicity.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi, folks;
We're running Postgres 7.3.2 and we have a core dump on HP-11.
This does not seem reproducible on Solaris or Linux.
Working with debugger we get this stack:

#0 0xc0185a20 in mallinfo+0x2144 () from /usr/lib/libc.2
(gdb) where
#0 0xc0185a20 in mallinfo+0x2144 () from /usr/lib/libc.2
#1 0xc01821c0 in __thread_callback_np+0x894 () from /usr/lib/libc.2
#2 0xc0187fa4 in realloc+0x1c4 () from /usr/lib/libc.2
#3 0xc018f420 in putenv+0xe8 () from /usr/lib/libc.2

It looks that problem occurs when we releasing TCL interpreter, also
the problem can be avoided if in pgtcl.c this code is commented out:

if (tclversion >= 8.1)
putenv("PGCLIENTENCODING=UNICODE");

I found another report on putenv problem:
http://archives.postgresql.org/pgsql-bugs/1999-11/msg00007.php

Are we're dealing with essentially same root cause?
Is there a recommended solution?

Thank you,
Michael.

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2003-06-23 20:53:22 Re: PlPython
Previous Message Dann Corbit 2003-06-23 20:34:11 Re: Two weeks to feature freeze