| From: | Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com> |
|---|---|
| To: | Vadim Ponomarev <vbponomarev(at)gmail(dot)com> |
| Cc: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> |
| Subject: | Re: Parallel heap vacuum |
| Date: | 2026-11-16 22:02:57 |
| Message-ID: | CALj2ACWZDLTRhKrD=u_J=ufV0xGz+cxa6r5W8=7yMbzmXFy-9Q@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
On Thu, Nov 12, 2026 at 1:13 PM Vadim Ponomarev <vbponomarev(at)gmail(dot)com> wrote:
>
> I have a question about the interaction between parallel heap scanning and the wraparound failsafe in v21.
>
> Is there a reason to avoid starting parallel heap workers when the failsafe is already active?
> For large tables parallel freezing could be particularly useful.
Thanks. I will address this in the next revision, but that's going to
be temporary. The final code depends on how the cutoffs are shared and
how it is detected by the leader and workers:
https://commitfest.postgresql.org/patch/7490/
--
Bharath Rupireddy
Amazon Web Services: https://aws.amazon.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bharath Rupireddy | 2026-11-16 22:05:00 | Re: Report oldest xmin source when autovacuum cannot remove tuples |
| Previous Message | Tristan Partin | 2026-11-16 22:02:05 | Re: Add malloc attribute to memory allocation functions |