From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Dave Perkins <drp(at)shore(dot)net> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: missing data |
Date: | 2001-03-29 20:30:08 |
Message-ID: | 24356.985897808@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Dave Perkins <drp(at)shore(dot)net> writes:
> Thanks for your time and patience . . . The only entries I see in the
> pg_class table are relnames with the 'pg_' prefix. They all have a
> relowner (40) that matches the only record listed in the pg_shadow
> table. The fact that the database in question does not have a
> corresponding entry in the pg_database seems to be most revealing if only I
> understood what it is revealing . . . The one other piece of evidence I
> have is the "Couldn't find any tables!" error message when executing the
> \dt command, yet when I run a select * for a relation that use to exist, it
> gives an acurate listing of field names, but no records (ie. missing data).
The "couldn't find any tables" complaint makes sense if pg_class isn't
showing any entries for user-defined tables. I don't quite understand
how it's letting you do a select from a table that's not visible in
pg_class, however.
Is it possible that you've executed more than 4 billion transactions
against this database? If so you might be suffering from ye infamous
XID wraparound problem. Look at pg_log in the $PGDATA directory; if
it's a gigabyte or so in size then this might have happened.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Dave Perkins | 2001-03-29 20:37:31 | Re: missing data |
Previous Message | Swaminathan Natarajan | 2001-03-29 20:19:35 | PLEASE HELP ME!: binary data in text/varchar? |