Re: (Never?) Kill Postmaster? - Mailing list pgsql-general

From Tommy Gildseth
Subject Re: (Never?) Kill Postmaster?
Date
Msg-id 471EE7EE.7060501@usit.uio.no
Whole thread Raw
In response to Re: (Never?) Kill Postmaster?  ("Pavel Stehule" <pavel.stehule@gmail.com>)
List pgsql-general
Pavel Stehule wrote:
> 2007/10/24, Stefan Schwarzer <stefan.schwarzer@grid.unep.ch>:
>
>> I read dozens of times the "TIP 2: Don't 'kill -9' the postmaster"...
>>
>> Now, what am I supposed to do if I launched a query which takes ages, and
>> which I want to interrupt?
> you have to use more gently way
>
> select pg_cancel_backend()
> http://www.postgresql.org/docs/current/interactive/functions-admin.html

The problem is that pg_cancel_backend() is somewhat unreliable at
cancelling wayward queries. You can try other options for kill though,
other than -9, which is kind of a last resort.

--
Tommy Gildseth


pgsql-general by date:

Previous
From: "Pavel Stehule"
Date:
Subject: Re: function result cache for pl/pgsql
Next
From: Christian Schröder
Date:
Subject: Re: (Never?) Kill Postmaster?