Madison Kelly wrote:
> Tom Lane wrote:
>> Madison Kelly <linux@alteeve.com> writes:
>>> PS - If I've run into a PgSQL bug, is there anything I can provide to help?
>>
>> A sequence that reproduces it would be the best thing ...
>
> I guess the trick is, I have no idea what's happened or what I did to
> cause it to happen... Any ideas I can try?
No ideas here ...
Can you please find out the current Xid counter? I think pg_controldata
should tell you.
Also, can you restore the previous state of pg_database, the one before
you deleted the old tuple? If not, maybe a pg_filedump of the table
file (in the hope that it hasn't been vacuumed) could show something
enlightening.
--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support