Re: Limit + group + join - Mailing list pgsql-performance

From Merlin Moncure
Subject Re: Limit + group + join
Date
Msg-id 6EE64EF3AB31D5448D0007DD34EEB3417DD207@Herge.rcsinc.local
Whole thread Raw
In response to Limit + group + join  (Tobias Brox <tobias@nordicbet.com>)
List pgsql-performance
Tobias wrote:
> Splendid :-) Unfortunately we will not be upgrading for some monthes
> still,
> but anyway I'm happy.  This provides yet another good argument for
> upgrading
> sooner.  I'm also happy to see such a perfect match:
>
>  - A problem that can be reduced from beeing complex and
>    production-specific, to simple and easily reproducible.
>
>  - Enthusiastic people testing it and pinpointing even more precisely
what
>    conditions will cause the condition
>
>  - Programmers actually fixing the issue
>
>  - Testers verifying that it was fixed
>
> Long live postgresql! :-)

In the last three or so years since I've been really active with
postgresql, I've found two or three issues/bugs which I was able to
reproduce and reduce to a test case.  In all instances the fix was in
cvs literally within minutes.

Merlin

pgsql-performance by date:

Previous
From: asif ali
Date:
Subject: Re: Weird performance drop after VACUUM
Next
From: "Marc G. Fournier"
Date:
Subject: getting an index to work with partial indices ...