Horton

Re: Intermittent buildfarm failures on wrasse

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, Noah Misch <noah@leadboat.com>, David Rowley <dgrowleyml@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-04-15T02:27:26Z
Lists: pgsql-hackers 
On Thu, Apr 14, 2022 at 7:20 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Oh!  You mean that maybe the OldestXmin horizon was fine, but something
> decided not to update hint bits (and therefore also not the all-visible
> bit) anyway?  Worth investigating I guess.

Yes. That is starting to seem like a plausible alternative explanation.

> > I'd really like to know what the removable cutoff
> > looks like for these VACUUM operations, which is something like
> > Andres' VACUUM VERBOSE debug patch should tell us.
>
> Yeah.  I'd hoped to investigate this manually and not have to clutter
> the main repo with debugging commits.

Suppose that the bug was actually in 06f5295af6, "Add single-item
cache when looking at topmost XID of a subtrans XID". Doesn't that fit
your timeline just as well?

I haven't really started to investigate that theory (just putting
dinner on here). Just a wild guess at this point.

-- 
Peter Geoghegan

Commits

  1. Repurpose PROC_COPYABLE_FLAGS as PROC_XMIN_FLAGS

  2. Tighten ComputeXidHorizons' handling of walsenders.

  3. Adjust VACUUM's removable cutoff log message.

  4. Temporarily add some probes of tenk1's relallvisible in create_index.sql.

  5. Set synchronous_commit=on in test_setup.sql.

  6. Rearrange core regression tests to reduce cross-script dependencies.

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