Horton

Re: error context for vacuum to include block number

Amit Kapila <amit.kapila16@gmail.com>

View thread View on postgr.es
From: Amit Kapila <amit.kapila16@gmail.com>
To: Justin Pryzby <pryzby@telsasoft.com>
Cc: Masahiko Sawada <masahiko.sawada@2ndquadrant.com>, Alvaro Herrera <alvherre@2ndquadrant.com>, Andres Freund <andres@anarazel.de>, Michael Paquier <michael@paquier.xyz>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2020-03-26T06:33:42Z
Lists: pgsql-hackers 
On Thu, Mar 26, 2020 at 10:11 AM Justin Pryzby <pryzby@telsasoft.com> wrote:
>
> Seems fine.  Rather than saying "different phases" I, would say:
> "The index vacuum and heap vacuum phases may be called multiple times in the
> middle of the heap scan phase."
>

Okay, I have slightly adjusted the wording as per your suggestion.

> But actually I think the concern is not that we unnecessarily "Revert back to
> the old phase" but that we do it in a *loop*.  Which I agree doesn't make
> sense, to go back and forth between "scanning heap" and "truncating".
>

Fair point.  I have moved the change to the truncate phase at the
caller of lazy_heap_truncate() which should address this concern.
Sawada-San, does this address your concern?

-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

Commits

  1. Avoid calls to RelationGetRelationName() and RelationGetNamespace() in

  2. Introduce vacuum errcontext to display additional information.

  3. Fix mesurement of elapsed time during truncating heap in VACUUM.

  4. Allow vacuum command to process indexes in parallel.

  5. Refactor code dedicated to index vacuuming in vacuumlazy.c

  6. Remove duplicated progress reporting during heap scan of VACUUM