| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com> |
| Cc: | henri(dot)chapelle(at)dbandmore(dot)com, pgsql-bugs(at)lists(dot)postgresql(dot)org |
| Subject: | Re: BUG #17942: vacuumdb doesn't populate extended statistics on partitioned tables |
| Date: | 2023-05-26 05:48:30 |
| Message-ID: | ZHBILpg7K4ylJ8zl@paquier.xyz |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
On Fri, May 26, 2023 at 09:15:36AM +0900, Kyotaro Horiguchi wrote:
> It seems to be exactly the same as ANALYZE, though. I'm a bit unclear
> about our perspective on this SQL command's behavior.
Sorry for being a bit unclear here. When dealing with partitioned
tables, a database-wide ANALYZE processes the partitions individually
as well as a full partition/inheritance tree.
My point is slightly different though: your suggestion of adding
RELKIND_PARTITIONED_TABLE to the filter added in vacuumdb would work
for -Z, but it would cause the vacuum code path of vacuumdb to process
more than once all the partitions in a single run. For instance, take
this schema:
CREATE TABLE parent_list (id int) PARTITION BY LIST (id);
CREATE TABLE child_list PARTITION OF parent_list
FOR VALUES IN (1, 2);
`vacuumdb` would now list both parent_list and child_list, making
child_list being vacuumed twice, which is not necessary. In order to
get a behavior in parity with the SQL commands ANALYZE, VACUUM and
VACUUM ANALYZE, we need to be more careful about the addition of
RELKIND_PARTITIONED_TABLE to the filtering clause.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kyotaro Horiguchi | 2023-05-26 07:49:35 | Re: BUG #17942: vacuumdb doesn't populate extended statistics on partitioned tables |
| Previous Message | Tim Field | 2023-05-26 01:36:14 | Comparing date strings with jsonpath expression |