| From: | Peter Geoghegan <pg(at)bowt(dot)ie> |
|---|---|
| To: | Mikhail Balayan <mv(dot)balayan(at)gmail(dot)com> |
| Cc: | Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Automatic aggressive vacuum on almost frozen table takes too long |
| Date: | 2023-02-17 21:34:57 |
| Message-ID: | CAH2-Wzn5WkfsyxAKMhEcqV+BdTFy1AfNSztw8mA1WrPYWzHu8w@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Thu, Feb 16, 2023 at 5:40 PM Mikhail Balayan <mv(dot)balayan(at)gmail(dot)com> wrote:
>> >> Do you have any non-btree indexes on the table? Can you show us the details of the
>> >> table, including all of its indexes? In other words, can you show "\d applications" output from psql?
>
> Only btree indexes. Please find the full table schema below:
It's possible that VACUUM had to wait a long time for a cleanup lock
on one individual heap page here, which could have added a long delay.
But...that doesn't seem particularly likely.
Can you run amcheck's bt_index_check() routine against some of the
indexes you've shown? There is perhaps some chance that index
corruption exists and causes VACUUM to take a very long time to delete
index pages. This is pretty much a wild guess, though.
--
Peter Geoghegan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ryan MYJ | 2023-02-18 00:00:18 | Hi All, |
| Previous Message | Arthur Ramsey | 2023-02-17 16:18:09 | Sequential scan faster than index |