Scott Marlowe <smarlowe(at)g2switchworks(dot)com> writes:
> I've noticed today that if one tries to pg_dump a database cluster
> running under pgpool, one gets the error message:
> pg_dump: query to get table columns failed: ERROR: kind mismatch
> between backends
> HINT: check data consistency between master and secondary
I would like to know exactly what pgpool has done to break pg_dump.
> Looking at the SQL that pg_dump sends to be relying on object OIDs for
> the dump.
There is nothing wrong with that technique ...
regards, tom lane
In response to
pgsql-general by date
|Next:||From: Scott Marlowe||Date: 2004-12-29 22:19:34|
|Subject: Re: pg_dump and pgpool|
|Previous:||From: Dave Smith||Date: 2004-12-29 21:58:58|
|Subject: Advice on moving rows to history|