John McCawley <nospam@hardgeus.com> writes:
> I am currently having a problem with a query never finishing (or at
> least not in a reasonable amount of time.) I have had similar problems
> to this in Postgres over the past several years, and I have always found
> workarounds. This time I'd like to actually understand what is happening.
If that join SELECT comes back in a reasonable period of time, then the
UPDATE shouldn't take too long either --- they're both doing about the
same thing as far as performing the join goes. The differential would
have to be index updates or triggers fired by the UPDATE. I'd bet on
the latter, but since you've told us zip about your schema or what PG
version this is, it's impossible to speculate further...
regards, tom lane