I am experiencing a similar problem with PostgreSQL running on Debian. I am not doing anything special when it
happens,just once and a while. It is not a production system (on which this problem does not occur) so I just restart
andeverything is fine.
I would be interested in seeing if this is happening to anyone else though.
Thanks,
Andrew
-----Original Message-----
From: pgsql-general-owner@postgresql.org
[mailto:pgsql-general-owner@postgresql.org]On Behalf Of Tom Lane
Sent: Tuesday, September 07, 2004 9:55 AM
To: Tore Halset
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] ERROR: canceling query due to user request
Tore Halset <halset@pvv.ntnu.no> writes:
> I am trying to port an old java application from MS SQL Server to
> PostgreSQL running on Mac OS X.
> ...
> The problem is that the server returns "ERROR: canceling query due to
> user request" on some of the queries.
The proximate cause of this has to be that something is sending SIGINT
to the backend process that's running the query. You'll have to look
around for reasons for that to happen, if you're sure that your client
code isn't doing it. A couple of possibilities come to mind:
1. You accidentally typed ^C on the terminal window the postmaster was
launched from.
2. The postmaster was launched under non-infinite resource limits
(ulimit settings) and whatever enforces that on OS X does it by sending
SIGINT rather than the more standard signals for such things. Do the
SIGINTs come at predictable times, such as when the backend has
accumulated X amount of runtime?
This isn't the first time I've heard of unexpected SIGINTs being
delivered on OS X. (I vaguely recall seeing the MySQL guys complaining
of just that.) I have not seen it happen in my own testing, though,
so I'm not sure what's going on. Let us know if you find the cause.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faqs/FAQ.html