| From: | Denis Perchine <dyp(at)perchine(dot)com> | 
|---|---|
| To: | Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp> | 
| Cc: | pgsql-general(at)postgresql(dot)org | 
| Subject: | Re: Problems during doing reindex on 7.0.3 | 
| Date: | 2000-11-14 05:19:48 | 
| Message-ID: | 00111411194809.08122@dyp.perchine.com | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-general | 
On 14 November 2000 07:33, Hiroshi Inoue wrote:
> Denis Perchine wrote:
> > Hello,
> >
> > I have very interesting problem. I have quite highly loaded database.
> > Once a day I create new indices, and after that drop old ones. Ido this
> > one by one.
>
> Your "reindex" is different from REINDEX command,isn't it ?
Sure. It is a sequnce like:
create index ix_name_1 ...;
drop index ix_name;
alter table ix_name_1 rename to ix_name;
For each index I like to recreate.
> > All is fine except this error message I sometimes get for some of the
> > queries just after reindex.
> >
> > ERROR:  Index 2050642 does not exist
> >
> > Looks like index oid was already resolved, but index was not locked
> > yet... And it was dropped just under the feet...
>
> Parser-rewriter-planner acquires a short term lock.
> This may be improved in 7.1 though I'm not sure.
That's really bad,  as it forces some of the queries to fail... And I can not 
detect the situation, as PostgreSQL does not have any classes for errors.
-- 
Sincerely Yours,
Denis Perchine
----------------------------------
E-Mail: dyp(at)perchine(dot)com
HomePage: http://www.perchine.com/dyp/
FidoNet: 2:5000/120.5
----------------------------------
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kevin O'Gorman | 2000-11-14 07:05:31 | Trying to understand VACUUM VERBOSE | 
| Previous Message | Tom Lane | 2000-11-14 03:44:28 | Re: GT.M database open sourced |