plperl db crash - Mailing list pgsql-novice

I had a db lock up on one of our machines.
To reproduce it, I did a fresh install on a testing machine.

On the other machine where it occured the first time the db just hung,
but it didn't crash. I tried to /etc/init.d/postgresql stop
But it didn't stop.

Then I did pg_ctl -m immediate stop
The log said something like 'redo is not required', so i gues everything
should be fine.

However, like already said, I did a fresh install on a testing machine
and there it crashes:

------------------------------------------------------------------------
debian/unstable
postgresql 7.3.4

/usr/lib/postgresql/bin/createlang plperl plasma
psql plasma

CREATE FUNCTION perlsub(text, text, text) RETURNS text AS '
     my ($data, $pat, $repl) = $_;
     $data =~ s/$pat/$repl/;
     return $data
' LANGUAGE 'plperl';

-- Taken from here:
-- http://archives.postgresql.org/pgsql-general/2001-02/msg00798.php

plasma=# SELECT perlsub('abc','b','123');
server closed the connection unexpectedly
         This probably means the server terminated abnormally
         before or while processing the request.
The connection to the server was lost. Attempting reset: Failed.
!#

---------------------------------------------------------------
Oct  2 19:15:00 billy postgres[23521]: [7] LOG:  connection received:
host=[local]
Oct  2 19:15:00 billy postgres[23521]: [8] LOG:  connection authorized:
user=plasma database=plasma
Oct  2 19:15:02 billy postgres[23329]: [7] LOG:  server process (pid
23521) was terminated by signal 11
Oct  2 19:15:02 billy postgres[23329]: [8] LOG:  terminating any other
active server processes
Oct  2 19:15:02 billy postgres[23329]: [9] LOG:  all server processes
terminated; reinitializing shared memory and semaphores
Oct  2 19:15:02 billy postgres[23523]: [10] LOG:  database system was
interrupted at 2003-10-02 19:13:01 CEST
Oct  2 19:15:02 billy postgres[23523]: [11] LOG:  checkpoint record is
at 0/83238C
Oct  2 19:15:02 billy postgres[23523]: [12] LOG:  redo record is at
0/83238C; undo record is at 0/0; shutdown TRUE
Oct  2 19:15:02 billy postgres[23523]: [13] LOG:  next transaction id:
513; next oid: 25168
Oct  2 19:15:02 billy postgres[23523]: [14] LOG:  database system was
not properly shut down; automatic recovery in progress
Oct  2 19:15:02 billy postgres[23523]: [15] LOG:  ReadRecord: record
with zero length at 0/8323CC
Oct  2 19:15:02 billy postgres[23523]: [16] LOG:  redo is not required
Oct  2 19:15:02 billy postgres[23524]: [10] LOG:  connection received:
host=[local]
Oct  2 19:15:02 billy postgres[23524]: [11] FATAL:  The database system
is starting up
Oct  2 19:15:04 billy postgres[23523]: [17] LOG:  database system is ready

---------------------------------------------------------------


--
  e-Trolley Sayegh & John, Nabil Sayegh
  Tel.: 0700 etrolley /// 0700 38765539
  Fax.: +49 69 8299381-8
  PGP : http://www.e-trolley.de