Horton

Re: Old row version in hot chain become visible after a freeze

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: "Wood, Dan" <hexpert@amazon.com>
Cc: "pgsql-bugs@postgresql.org" <pgsql-bugs@postgresql.org>
Date: 2017-08-31T22:56:20Z
Lists: pgsql-bugs 
Hi Dan,

Nice to hear from you.

On Thu, Aug 31, 2017 at 3:36 PM, Wood, Dan <hexpert@amazon.com> wrote:
> Because tupgone is false we freeze instead of deleting.  Freezing a DEAD
> tuple makes it visible.  Here is a comment in heap_prepare_freeze_tuple()
>
>
>
>   * It is assumed that the caller has checked the tuple with
>
>   * HeapTupleSatisfiesVacuum() and determined that it is not HEAPTUPLE_DEAD

Funny that there'd be another bug associated with
heap_prepare_freeze_tuple() so soon after the last one was discovered.
Are you aware of the other one [1]?

BTW, I just posted a patch to enhance amcheck, to allow it to verify
that an index has all the entries that it ought to [2]. Perhaps you'd
find it useful for this kind of thing. I welcome your feedback on
that.

[1] https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=31b8db8e6c1fa4436116f4be5ca789f3a01b9ebf;hp=f1dae097f2945ffcb59a9f236843e0e0bbf0920d
[2] https://commitfest.postgresql.org/14/1263/
-- 
Peter Geoghegan

Commits

  1. Fix freezing of a dead HOT-updated tuple

  2. Fix potential data corruption during freeze

  3. Clarify the contract of partition_rbound_cmp().