Horton

Re: [CLOBBER_CACHE]Server crashed with segfault 11 while executing clusterdb

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

View thread View on postgr.es
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Amul Sul <sulamul@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, Michael Paquier <michael@paquier.xyz>, Noah Misch <noah@leadboat.com>, Neha Sharma <neha.sharma@enterprisedb.com>, Amit Langote <amitlangote09@gmail.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2021-07-09T14:00:34Z
Lists: pgsql-hackers 
On 2021-Jul-09, Amul Sul wrote:

> > On Tue, Jul 6, 2021 at 11:06 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:

> > > The point of the static-inline function idea was to be cheap enough
> > > that it isn't worth worrying about this sort of risky optimization.
> > > Given that an smgr function is sure to involve some kernel calls,
> > > I doubt it's worth sweating over an extra test-and-branch beforehand.
> > > So where I was hoping to get to is that smgr objects are *only*
> > > referenced by RelationGetSmgr() calls and nobody ever keeps any
> > > other pointers to them across any non-smgr operations.

> Herewith attached version did the same, thanks.

I think it would be valuable to have a comment in that function to point
out what is the function there for.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/
"Industry suffers from the managerial dogma that for the sake of stability
and continuity, the company should be independent of the competence of
individual employees."                                      (E. Dijkstra)

Commits

  1. Replace RelationOpenSmgr() with RelationGetSmgr().

  2. Avoid possible crash while finishing up a heap rewrite.

  3. Redefine pg_class.reltuples to be -1 before the first VACUUM or ANALYZE.