Re: Detecting mis-planning of repeated application of a projection step

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "pbj(at)cmicdo(dot)com" <pbj(at)cmicdo(dot)com>
Cc: "pgsql-general(at)lists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org>
Subject: Re: Detecting mis-planning of repeated application of a projection step
Date: 2021-10-13 15:42:32
Message-ID: 653389.1634139752@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

"pbj(at)cmicdo(dot)com" <pbj(at)cmicdo(dot)com> writes:
> The 13.4 release notes document the following:
>  > Fix mis-planning of repeated application of a projection step (Tom Lane)
>  >
>  >      The planner could create an incorrect plan in cases where two
>  >      ProjectionPaths were stacked on top of each other. The only known
>  >      way to trigger that situation involves parallel sort operations,
>  >      but there may be other instances. The result would be crashes or
>  >      incorrect query results. Disclosure of server memory contents is also
>  >      possible. (CVE-2021-3677)
>  >

> Is there a way to detect if this is happening and could this problem cause very bad, but successful, plans to be created?  We are seeing 13.3 create horrible plans for some queries, where 13.4 creates great plans for the same queries.

No, whatever you're seeing, it was not this particular bug.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Phil Endecott 2021-10-13 16:09:36 Replication between different architectures
Previous Message pbj@cmicdo.com 2021-10-13 15:34:57 Detecting mis-planning of repeated application of a projection step