Re: Restarting standby database have to reaply WAL (redo logs)

On Tue, 2007-09-25 at 14:14 -0300, Cesar Otero wrote:

> I think that after restarting standby db the first log should be last
> log applied + 1

That isn't how it works, for performance reasons.

We don't make a restartpoint after each file, we do this on the first
usable checkpoint WAL record that we find after checkpoint_timeout
seconds.

--
  Simon Riggs
  2ndQuadrant  http://www.2ndQuadrant.com