On Tue, Jan 7, 2020 at 12:47 PM David Fetter wrote:
> Per a suggestion Christophe made, please find attached a patch to
> $Subject:
Curious, what's the benefit of autovacuum handling the oldest tables
first? If there is a related thread with the discussion, I couldn't find
it.
> Apart from caref
On Thu, Jan 9, 2020 at 12:07 PM Robert Haas wrote:
>
> On Thu, Dec 12, 2019 at 2:26 PM David Fetter wrote:
> > > I wonder if you might add information about table size, table changes,
> > > and bloat to your RelFrozenXidAge struct and modify rfxa_comparator to
> > > use a heuristic to cost the (a
Hi Hackers,
Is it fair to assume that, given the same data, a partitioned table should
return the same results as a non-partitioned table? If that's true, then I
think I may have stumbled across a case of wrong results on boolean partitioned
tables.
In following example, I think we incorrectly s
On Wed, Apr 12, 2023 at 4:13 AM David Rowley wrote:
> On Wed, 12 Apr 2023 at 22:13, David Kimura wrote:
> > Is it fair to assume that, given the same data, a partitioned table should
> > return the same results as a non-partitioned table?
>
> Yes, and
On Wed, Apr 12, 2023 at 4:13 AM David Rowley wrote:
>
> There seems to be a bunch of tests checking this already, all of them
> assuming the incorrect plans.
Given that the plan alone wasn't sufficient to catch this error previously,
would it be worthwhile to add some data to the tests to make it
e approach I had
considered was to update the rewriteHandler to explicitly force the existence of
the duplicate alias column in the range tables. But that also felt meh.
Does this seem like a legit issue? And if so, any thoughts on alternative
approaches?
Thanks,
Dav
On Wed, Apr 29, 2020 at 4:39 PM Melanie Plageman
wrote:
>
> In addition to many assorted TODOs in the code, there are a few major
> projects left:
> - Batch 0 falling back
> - Stripe barrier deadlock
> - Performance improvements and testing
>
Batch 0 never spills. That behavior is an artifact of
On Wed, Apr 29, 2020 at 4:44 PM David Kimura wrote:
>
> Following patch adds logic to create a batch 0 file for serial hash join so
> that even in pathalogical case we do not need to exceed work_mem.
Updated the patch to spill batch 0 tuples after it is marked as fallback.
A couple
