From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Adam Bultman <adam(dot)bultman(at)iconideas(dot)com> |
Cc: | pgsql-novice(at)postgresql(dot)org |
Subject: | Re: Killing postgres processes. |
Date: | 2002-07-24 20:38:30 |
Message-ID: | 13690.1027543110@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-novice |
Adam Bultman <adam(dot)bultman(at)iconideas(dot)com> writes:
> Greetings. I've got a postgres server that's being hit rather heavily.
> I have many processes inititated by users (they start scripts running).
> Because of user impatience and idiocy, despite my explanations, they
> continue to re-run the script and not wait for the script to end the
> first time (i.e. 'why isn't it finishing? <click click click>'). So, I
> have about 9 postgres SELECTS going on right now that are pretty much
> runaway and useless. Can I safely kill these without fubaring things?
Yes. Best bet is to send them SIGINT, which will cause a clean query
cancel, and then they'll notice the client connection went away and
quit. More drastic measures might fool the postmaster into thinking
that a backend crashed, in which case it'll force a database-wide
restart, which you probably don't want.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Jesús López | 2002-07-24 23:29:24 | extract data and store |
Previous Message | Josh Berkus | 2002-07-24 17:47:51 | Re: maximum size of psql query file |