There is also the statement_timeout setting in postgresql.conf, but
you have to be careful with this setting. I'm not sure about
postgres 8.0 or 8.1, but in 7.4.5 this setting will terminate the
COPY statements used by pg_dumpall for backups. So I actually use
the pg_stat_activity table t
Hi,
On Tue, 2006-05-02 at 17:19 -0600, Dan Harris wrote:
> Is there some way I can just kill a query and not risk breaking
> everything else when I do it?
Use pg_stat_activity view to find the pid of the process (pidproc
column) and send the signal to that process. I think you are now killing
po
Tom Lane wrote
You should be using SIGINT, not SIGTERM.
regards, tom lane
Thank you very much for clarifying this point! It works :)
---(end of broadcast)---
TIP 6: explain analyze is your friend
On 5/2/06, Dan Harris <[EMAIL PROTECTED]> wrote:
My database is used primarily in an OLAP-type environment. Sometimes my
users get a little carried away and find some way to slip past the
sanity filters in the applications and end up bogging down the server
with queries that run for hours and ho
Dan Harris <[EMAIL PROTECTED]> writes:
> So, I have been searching for a way to kill an individual query. I read
> in the mailing list archives that you could 'kill' the pid. I've tried
> this a few times and more than once, it has caused the postmaster to
> die(!), terminating every query tha
My database is used primarily in an OLAP-type environment. Sometimes my
users get a little carried away and find some way to slip past the
sanity filters in the applications and end up bogging down the server
with queries that run for hours and hours. And, of course, what users
tend to do is
