Simon Riggs writes:
> On Wed, Apr 20, 2011 at 7:41 PM, Tom Lane wrote:
>> It's still true though that you have to be REINDEXing system catalogs to
>> be at risk, else you shouldn't be seeing any IN_PROGRESS tuples.
> So the fix seems to be that we make REINDEX on a system catalog lock
> the whol
On Wed, Apr 20, 2011 at 7:41 PM, Tom Lane wrote:
> It's still true though that you have to be REINDEXing system catalogs to
> be at risk, else you shouldn't be seeing any IN_PROGRESS tuples.
So the fix seems to be that we make REINDEX on a system catalog lock
the whole catalog table.
Anything e
What with the recent discussions, I've been looking harder at the
REINDEX code's interactions with HOT, and I've found another problem
altogether. To wit, IndexBuildHeapScan considers the DELETE_IN_PROGRESS
case to be comparable to RECENTLY_DEAD, but that analogy fails for
HOT-updated tuples. If
