Horton

Re: pgsql: reindexdb: Add the index-level REINDEX with multiple jobs

Alvaro Herrera <alvherre@alvh.no-ip.org>

View thread View on postgr.es
From: Álvaro Herrera <alvherre@alvh.no-ip.org>
To: Alexander Korotkov <aekorotkov@gmail.com>
Cc: Alexander Korotkov <akorotkov@postgresql.org>, pgsql-hackers@lists.postgresql.org, Maxim Orlov <orlovmg@gmail.com>, Svetlana Derevyanko <s.derevyanko@postgrespro.ru>
Date: 2025-03-12T09:41:43Z
Lists: pgsql-hackers 
On 2025-Mar-11, Alexander Korotkov wrote:

> Thank you for your feedback!  I also think that pipelining would be a
> better options, but it's too invasive for backpatching. I've written
> comments for gen_reindex_command() and run_reindex_command().  I'm
> going to push this patch to master and 17 if no objections.

Looks good.

Actually, I don't understand why is run_reindex_command() calling
PQfinish().  Hmm, also, why is the 'async' parameter always given as
true?  Is that PQfinish() actually dead code?

[... Looks at git history ...]

Ah!  this is actually a problem older than your patch -- it appears that
the last (only?) caller with async=false was removed by commit
2cbc3c17a5c1.  I think it's worth removing that while you're at it.  We
no longer need executeMaintenanceCommand() in reindexdb.c anymore
either, I think.

-- 
Álvaro Herrera        Breisgau, Deutschland  —  https://www.EnterpriseDB.com/

Commits

  1. reindexdb: Fix the index-level REINDEX with multiple jobs

  2. reindexdb: Fix warning about uninitialized indices_tables_cell

  3. reindexdb: Add the index-level REINDEX with multiple jobs

  4. Rework logic and simplify syntax of REINDEX DATABASE/SYSTEM