Re: Bug #474: Index using problem - Mailing list pgsql-bugs

From Tom Lane
Subject Re: Bug #474: Index using problem
Date
Msg-id 21296.1002309306@sss.pgh.pa.us
Whole thread Raw
In response to Re: Bug #474: Index using problem  (Andreas Wernitznig <andreas@insilico.com>)
Responses Re: Bug #474: Index using problem  (Andreas Wernitznig <andreas@insilico.com>)
List pgsql-bugs
Andreas Wernitznig <andreas@insilico.com> writes:
> -> Using one connection the optimizer for pk/fk-checking is not
> updated by a "vacuum analyze".

Oh, I misunderstood you the first time: I thought you were saying that
*other* backends couldn't see the results of the VACUUM.

The reason for this behavior is that the foreign key checker caches a
plan for each foreign-key-checking query the first time it needs to
use that query (within a given backend).  There should be a mechanism
to flush those cached plans when circumstances change ... but currently
there isn't.

            regards, tom lane

pgsql-bugs by date:

Previous
From: Andreas Wernitznig
Date:
Subject: Re: Bug #474: Index using problem
Next
From: Andreas Wernitznig
Date:
Subject: Re: Bug #474: Index using problem