| From: | Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at> |
|---|---|
| To: | Peter Geoghegan <pg(at)bowt(dot)ie>, pgsql-committers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: pgsql: Drop index behind pg_upgrade test issue. |
| Date: | 2021-03-11 09:36:37 |
| Message-ID: | 9edc8c330b9a5e3426e8cf57fc5649419eada03b.camel@cybertec.at |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-committers |
On Thu, 2021-03-11 at 03:02 +0000, Peter Geoghegan wrote:
> The vacuum_cleanup_index_scale_factor storage parameter was set in a
> btree index that was previously left behind in the regression test
> database. As a result, the index gets tested within pg_dump and
> pg_restore tests, as well as pg_upgrade testing. This won't work when
> upgrading to Postgres 14, though, because the storage parameter was
> removed on that version by commit 9f3665fb.
>
> Fix the test failure by dropping the index in question.
I'd take that as an indication that Michael had a point with his
complaint[1]. Perhaps it would be better to implement one of his
ideas than to remove the test.
Yours,
Laurenz Albe
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alvaro Herrera | 2021-03-11 16:13:01 | Re: pgsql: Don't consider newly inserted tuples in nbtree VACUUM. |
| Previous Message | Michael Paquier | 2021-03-11 08:16:37 | pgsql: Set libcrypto callbacks for all connection threads in libpq |