Ludwig Isaac Lim <[EMAIL PROTECTED]> writes:
> Under normal circumstances, there's no need to vacuum
> template1 right?
Only if it hasn't been modified.
> Its quite difficult to imagine XID
> wraparound problems occuring in template1. I can't think
> how can 4 billion transactions occur in template1 if
> template1 is just used as a template in creating a new
> database.
> Am I missing something here?
Yes: the wraparound limit is for transactions cluster-wide, not per
database. If you make 1 change in template1 and then 2 billion
changes in some other databases, template1 is broken unless it's
been vacuumed meanwhile.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend
