On Sep 15, 2012, at 11:32 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Right, but we do a shutdown checkpoint at the end of crash recovery.
Yes, but that only writes the buffers that are dirty. It doesn't fix the lack of a BM_PERMANENT flag on a buffer that
oughtto have had one. So that page can now get modified AGAIN, after recovery, and not checkpointed.
> I could believe that this case gets missed, but it's not clear from
> what's been said --- and if that case *is* broken, there should have
> been a whole lot more corruption reports recently than what we've seen.
It's pretty clear from Jeff's example at the top of the thread, which involves two crashes and no archive recovery.
...Robert
