Horton

Re: XLogReadRecord() error in XlogReadTwoPhaseData()

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Noah Misch <noah@leadboat.com>
Cc: Andres Freund <andres@anarazel.de>, pgsql-hackers@lists.postgresql.org, Thomas Munro <thomas.munro@gmail.com>, pgbf@twiska.com, Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2022-01-24T02:20:21Z
Lists: pgsql-hackers 
On Sun, Jan 23, 2022 at 06:10:07PM -0800, Noah Misch wrote:
> Could do that.  Every run that doesn't get the flaky failure will print a
> message like "TODO passed:  3-5", though the test file could mitigate that by
> declaring the TODO only on configurations where we expect a failure.  The
> 027_stream_regress.pl trouble involves reaching a die(), not failing a test in
> this sense, so that one would take more work.

Using a TODO has the advantage to allow the tests to run on a periodic
basis, even if they could fail in this unexpected way.  We've had our
load of issues in the past proper to specific architectures, so this
could help with the 2PC code paths and SPARC.  Not to mention that we
could just let the code be as-is, and it would fix itself once the
kernel is updated.  So that sounds much better to me than a skip
phase.
--
Michael

Commits

  1. Under has_wal_read_bug, skip recovery/t/032_relfilenode_reuse.pl.

  2. Under has_wal_read_bug, skip contrib/bloom/t/001_wal.pl.

  3. Use Test::Builder::todo_start(), replacing $::TODO.

  4. On sparc64+ext4, suppress test failures from known WAL read failure.

  5. Report any XLogReadRecord() error in XlogReadTwoPhaseData().