Re: Parallel autovacuum: leader crashes when no DSM segment can be created

From: shihao zhong <zhong950419(at)gmail(dot)com>
To: Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com>
Cc: Ayush Tiwari <ayushtiwari(dot)slg01(at)gmail(dot)com>, Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Parallel autovacuum: leader crashes when no DSM segment can be created
Date: 2026-10-30 02:18:57
Message-ID: CAGRkXqQHk9VjvXyRwTvOnJBv1K276uXCyDLb0kDvNjSHFFMeAQ@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

Bharath is right that it still fails when only one slot is left. But
I think v1 still fixes the case where all slots are taken.

That said, running out of DSM slots seems very rare. I have never seen
it happen, and this thread is the first time I learned the limit
exists.

If we think it is a real problem, how about reserving some DSM slots
just for vacuum? That would be similar to how autovacuum_worker_slots
gives autovacuum workers their own pool.

Thanks,
Shihao

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message shihao zhong 2026-10-30 02:27:21 Re: [PG19] pg_upgrade --check passes, then the upgrade fails
Previous Message Tom Lane 2026-10-30 02:12:07 Re: Return INT_MAX or INT_MIN from strtoint() on overflow