> I am not quite certain that this is completely bulletproof when there
> are multiple backends concurrently updating pg_shadow, but I have not
> been able to think of a case where it'd fail. The worst possibility
> is that a committed update in pg_shadow might not get propagated to
> pg_pwd for a while because some other transaction is holding a lock on
> pg_shadow. (But pg_pwd updates can be delayed for that reason now,
> so it's certainly no worse than before.)
>
> Comments?
I see your point. Guess we have to do it inside xact.
--
Bruce Momjian | http://www.op.net/~candle
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026