From: | Bruce Momjian <maillist(at)candle(dot)pha(dot)pa(dot)us> |
---|---|
To: | raines(at)SLAC(dot)Stanford(dot)EDU |
Cc: | hackers(at)postgreSQL(dot)org (PostgreSQL-development), vadim(at)sable(dot)krasnoyarsk(dot)su (Vadim B(dot) Mikheev) |
Subject: | Re: Let's talk up 6.3 |
Date: | 1998-03-28 21:30:20 |
Message-ID: | 199803282130.QAA21693@candle.pha.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
>
> >> Subselects are a BIG item for 6.3, and this is a serious feature that we
> >> should be telling people about. In the past, I am sure certain people
> >> did not consider using PostgreSQL because of this missing feature.
> >>
>
> Yes, they are a big reason I want to use PostgreSQL, but as far as
> I can tell, they do not work. Is there a patch I am missing?
> I have 6.3.1 on RedHat Linux 5.0.
Nope, this is the first problem I have heard about with subselects.
>
> Here is what I tried:
> ======================================================================
> bbrmdc=> select runnum from mdc1_simu where version = '4.3.7g';
> runnum
> ------
> 048930
> 048931
> 048932
> 048933
> 048934
> (5 rows)
>
> bbrmdc=> select distinct runtype from mdc1_runs where runnum in
> bbrmdc-> ('048930','048931','048932','048933','048934');
> runtype
> --------------------
> tau+ -> X, tau- -> X
> (1 row)
>
> bbrmdc=> select distinct runtype from mdc1_runs where runnum in
> bbrmdc-> (select runnum from mdc1_simu where version = '4.3.7g');
> FATAL: unrecognized data from the backend. It probably dumped core.
> FATAL: unrecognized data from the backend. It probably dumped core.
> bbrmdc=> \q
>
> ======================================================================
>
> Each of the single selects took < 1 sec. The fatals are that after 15
> minutes, I killed the postgres process on my server. BTW, is there
> clean way to kill a query from the psql side? Doing a Ctrl-C just
> kills the psql process and leaves the postgres process eating up my
> CPU on the server.
No way to cancel them, but it is on the TODO list.
I am CC'ing Vadim on this. Looks strange. Any way we can reproduce
this? Does the removal of the DISTINCT help? Are there a lot of values
without the DISTINCT?
--
Bruce Momjian | 830 Blythe Avenue
maillist(at)candle(dot)pha(dot)pa(dot)us | Drexel Hill, Pennsylvania 19026
+ If your life is a hard drive, | (610) 353-9879(w)
+ Christ can be your backup. | (610) 853-3000(h)
From | Date | Subject | |
---|---|---|---|
Next Message | Paul Raines | 1998-03-28 23:04:25 | Re: Let's talk up 6.3 |
Previous Message | Bruce Momjian | 1998-03-28 21:22:44 | Re: [PORTS] Port Bug Report: pg_dump -d database >unload.file; cat unload.file|psql database ARE NOT EQUAL |