On Wed, May 25, 2011 at 6:37 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
> On Wed, May 25, 2011 at 1:08 PM, Andrew Sullivan <ajs@crankycanuck.ca> wrote:
>
>> Note that "unneeded". Obviously, you need more than that, probably
>> because you're restoring the database in one transaction (so none of
>> the files can be flushed).
>
> That's the way SQLServer and Oracle work, but not PostgreSQL. We can
> clear down WAL files even during a long running transaction.
>
> For us, "unneeded" means prior to the second-to-last checkpoint record.
I wonder if OP is outrunning his checkpoint writing?