Yes, it'll core dump if select * from another big table. Here're some extra
info -
emrxdbs=# explain select * from patient;
QUERY PLAN
----------------------------------------------------------------------------
---
Seq Scan on patient (cost=100000000.00..100083605.77 rows=3296977
width=147)
(1 row)
emrxdbs=# explain analyze select * from patient;
QUERY PLAN
----------------------------------------------------------------------------
-------------------------------------------------------
Seq Scan on patient (cost=100000000.00..100083605.77 rows=3296977
width=147) (actual time=0.052..18093.297 rows=3530593 loops=1)
Total runtime: 21877.238 ms
(2 rows)
emrxdbs=# select * from patient;
Illegal instruction (core dumped)
Thanks.
johnl
-----Original Message-----
From: Scott Marlowe [mailto:smarlowe@qwest.net]
Sent: Monday, June 07, 2004 4:31 PM
To: John Liu
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] core dump
On Mon, 2004-06-07 at 15:57, John Liu wrote:
> We upgraded from 7.2 to 7.4, it looks like everything working, but
> when I issue a query such as select * from tab (tab has about 2-3
> million records), it causes core dump. I tuned some the parameters, it
> still produce the core.
How odd. Anything else cause it to dump quickly?
Does it core if you just run explain select ... (note the lack of an
analyze there, as we don't want anything but the planner to run.)
Does selecting from another large table cause the same problem, or is
this a one table thing?