Brent Friedman skrev:
> I tried rewriting an existing view based on a multi-way join into
> several smaller views, to keep the RAM footprint of a reporting query
> as small as possible. One of these views (in a series) is being
> corrupted/changed by postgres.
> SELECT DISTINCT vw2.other_table_id, vw2.somedata1, vw2.somedata2,
> vw2.somedata3, vw2.somedata4,
> ORDER BY vw2.station_transactionid, vw2.accepted, vw2.rejected, vw2.bulk,
Did you by any chance forget to anonymize the field names in the last line?
In response to
pgsql-novice by date
|Next:||From: Josh Berkus||Date: 2007-10-01 17:13:41|
|Subject: Re: [HACKERS] PG on NFS may be just a bad idea|
|Previous:||From: Zdenek Kotala||Date: 2007-09-29 14:23:59|
|Subject: Re: PG on NFS may be just a bad idea|