From: | Alex Krohn <alex(at)gossamer-threads(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: select fails on indexed varchars. |
Date: | 2001-01-27 00:03:49 |
Message-ID: | 20010126160035.0BF2.ALEX@gossamer-threads.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
Hi,
> > So I added:
> > LANG=C
> > LC_ALL=C
> > to the /etc/rc.d/init.d/postgres file and stop, restarted the server. I
> > then dropped and recreated the database. However still same results.
>
> LC_COLLATE overrides LC_ALL, I think --- didn't you previously show us
> that all the LC_xxx family variables were set in your default
> environment? You may need to set (or unset if you prefer) all of 'em.
>
> Also, I'd really recommend an initdb, not the above half-baked approach,
> because the above will not fix any problems that the template1 indexes
> might have with a changed sort order.
I added to the startup file:
LANG=C
LC_CTYPE=C
LC_NUMERIC=C
LC_TIME=C
LC_COLLATE=C
LC_MONETARY=C
LC_MESSAGES=C
LC_ALL=C
as well as to the postgres users default environment. I then shut down
postmaster, and as user postgres ran `initdb /var/lib/pgsql`. I then ran
/etc/rc.d/init.d/postgres start as root, and then as user postgres ran
`createdb mytest`.
After this, my create test and select still produced the same error. Ugh.
Cheers,
Alex
From | Date | Subject | |
---|---|---|---|
Next Message | Robert B. Easter | 2001-01-27 00:31:35 | Re: BUG in postgres mathematic |
Previous Message | Barry Lind | 2001-01-26 23:37:50 | Re: [BUGS] no way in LargeObject API to detect short read? |