Simon Riggs wrote:
> On Thu, 2009-10-22 at 07:55 +0300, Heikki Linnakangas wrote:
>> Making some effort to transfer locks instead of acquiring+releasing
>> would eliminate the need for having extra lock space available when
>> switching from hot standby mode to normal operation.
>
> This isn't very clear. You started by saying you were quite eager to
> always grant and then release; this sounds like you don't want that now,
> but you now again like the approach I had already attempted to take.
Yeah, I haven't made up my mind. What's in there now is certainly
broken, so we need to do something. The simplest approach would be to
revert the changes in lock_twophase_recover(), while transfering the
locks with something like AtPrepare_Locks() would be more robust in the
face of shared memory shortage.
-- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com