On Thu, 2009-06-25 at 15:10 -0400, Tom Lane wrote:
> So to summarize the state of play, it seems
> we have these issues:
>
> * need to delete startup process's local pendingOpsTable once bgwriter
> is launched, so that requests go to bgwriter instead
Need to ensure that fsync requests are directed to the process that will
act on the fsync requests.
> * need to push end-of-recovery checkpoint into bgwriter
That's probably the easiest thing to do, but the issue is that we must
fsync all files mentioned in the pendingOpsTable in *any* process that
has been accumulating such requests.
> * need to do something about IsRecovery tests that will now be executed
> in bgwriter
Yes
> * need to fix mistaken code in FinishPreparedTransaction
Yes
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support