"Paul B. Anderson" <paul.a@pnlassociates.com> writes:
> I have a cluster of machines and the databases are on shared disk
> storage. I'm just getting this arrangement working and, while I've been
> debugging my scripts, I've accidentally had two copies of postgresql
> running against the same initdb directories at the same time on two
> different machines. Needless to say, this causes problems and I'm
> thinking that vacuuming in this condition may be at or near the root of
> the problem.
You're going to have much worse problems than this from that. Two
postmasters modifying the same data files is a recipe for complete
corruption. I would recommend abandoning any database you realize
this has happened to: just initdb and start over. The chances of
being able to fix it are minimal.
regards, tom lane