I'm interested in point-in-time-recovery (or archive recovery) too.
I'm wondering current implementation of CHECKPOINT is whether "SHARP"
or "FUZZY" checkopint. If it's a "SHARP" one, it would be a serious
performance bottle neck according to the Gray's transaction book. Does
anybody know anything about that?
pgsql-hackers by date
|Next:||From: Bruce Momjian||Date: 2002-02-24 03:58:25|
|Subject: Duration of beta period|
|Previous:||From: Oliver Elphick||Date: 2002-02-23 21:52:44|
|Subject: Re: pg_ctl - tighten command parameter checking|