Horton

Re: BUG #17255: Server crashes in index_delete_sort_cmp() due to race condition with vacuum

Andres Freund <andres@anarazel.de>

View thread View on postgr.es
From: Andres Freund <andres@anarazel.de>
To: Peter Geoghegan <pg@bowt.ie>
Cc: Dmitry Dolgov <9erthalion6@gmail.com>, Alexander Lakhin <exclusion@gmail.com>, Matthias van de Meent <boekewurm+postgres@gmail.com>, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>
Date: 2022-03-11T04:08:08Z
Lists: pgsql-bugs 
Hi,

On 2022-03-10 19:52:24 -0800, Peter Geoghegan wrote:
> As the patch points out, we are required to establish rel_pages after
> we have established OldestXmin.

Hm. Now that I think more about it, I'm not as convinced that's entirely true
anymore. We just need to make sure that we have an accurate "xid horizon" from
before determining the relation size, not an "xmin horizon" and use the Min()
of the two horizons.  The "xid horizon" obviously is typically well ahead of
the "xmin horizon" (and always >=).

And even on the "xid horizon" I suspect something more aggressive could be
used.

But that's probably best done separately - but perhaps worth to "weaken" the
comment a bit?

Greetings,

Andres Freund

Commits

  1. vacuumlazy.c: Standardize rel_pages terminology.

  2. vacuumlazy.c: document vistest and OldestXmin.

  3. heap pruning: Only call BufferGetBlockNumber() once.

  4. Fix possible HOT corruption when RECENTLY_DEAD changes to DEAD while pruning.

  5. Assert redirect pointers are sensible after heap_page_prune().

  6. Remove tupgone special case from vacuumlazy.c.

  7. Recycle nbtree pages deleted during same VACUUM.

  8. snapshot scalability: Don't compute global horizons while building snapshots.

  9. Refactor heap_page_prune so that instead of changing item states on-the-fly,