[rt-users] Performance problem with SG-clusteren and postgres 8.0.7

[Ruslan Zakirov]

This is one of the most problematic query in RT. I only can confirm
that the first plan (on Pg 7.x) is almost optimal one (there is still
way to optimize it with planner) and this is plan we wanted to achive
when were changing this query in RT-3.4.5.

I hope Pg 8.1.4 would be available soon and we could recommend that
instead of Pg 8.0.x.

Thanks to you and your team for the report and help.

On 4/10/06, Tomas Olaj <tomas.olaj at usit.uio.no> wrote:
>
> Just translating a message from our DBA Rafael Martinez,
>
> -->
>
> We have a huge problem ahead of us which may delay our Service Guard
> Cluster project.
>
> Our last performance test with our new cluster and RT shows that we
> have a SQL statement that 'planner" in Postgres 8.0.7 doesn't manage to
> find a good 'plan' for and takes to long to execute.
>
> I have sent an e-mail to pgsql-performance and got a reply back, and a
> solution that implies to upgrade to 8.1.4 when it's shipped.
>
> The e-mail I sent (including the SQL) to pgsql-performance:
> http://archives.postgresql.org/pgsql-performance/2006-04/msg00125.php
>
> Answers from Tom Lane (one of the major developers in the PostgreSQL project):
> http://archives.postgresql.org/pgsql-performance/2006-04/msg00138.php
> http://archives.postgresql.org/pgsql-performance/2006-04/msg00142.php
>
> --
> Rafael Martinez, <r.m.guerrero at usit.uio.no>
> Center for Information Technology Services
> University of Oslo, Norway
>
> PGP Public Key: http://folk.uio.no/rafael/
>
> -->
>
> Tomas, RT management
>
> --
> ________________________________________________________________________
> Tomas A. P. Olaj, email: tomas.olaj at usit.uio.no, web: folk.uio.no/tomaso
>   University of Oslo / USIT (Center for Information Technology Services)
>     System- and Application Management / Applications Management Group


--
Best regards, Ruslan.