Re: Progress reporting: a debug trace and a test framework

From: Manu <manuelreyesbravo(at)gmail(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Cc: Álvaro Herrera <alvherre(at)kurilemu(dot)de>, Michael Paquier <michael(at)paquier(dot)xyz>, Fujii Masao <masao(dot)fujii(at)oss(dot)nttdata(dot)com>
Subject: Re: Progress reporting: a debug trace and a test framework
Date: 2026-11-20 14:43:29
Message-ID: 179518580956.1274549.1607725125235375604@gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

v7, attached, is rebased on master, where the previous version no
longer passes: 51ae21aa313 added two setup phases to REPACK
(CONCURRENTLY), and with PROGRESS_DEBUG, 001 failed on both REPACK
(CONCURRENTLY) phase subtests. That is the kind of change the test
is there to catch.

001 now expects the two phases. The first, "enabling logical
decoding", is optional: it has nothing to do when effective_wal_level
is already logical, and the backend may wake up after it is over. In
20 runs of 001 with the CPUs kept busy, 8 of the 40 REPACK
(CONCURRENTLY) runs reported it and 32 did not, and all 20 runs
passed. Nothing else changed.

check-world passes without PROGRESS_DEBUG, with it, and with it plus
trace_progress on in every test server.

Regards,
Manu

Attachment Content-Type Size
v7-0001-Add-PROGRESS_DEBUG-to-test-the-whole-sequence-of-.patch text/x-patch 118.5 KB
v7-0002-Check-the-documented-progress-phases-against-the-.patch text/x-patch 12.0 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Greg Burd 2026-11-20 15:01:46 Re: Hash index: stop registering pin-only bucket pages in WAL (remove REGBUF_NO_CHANGE)
Previous Message Alvaro Herrera 2026-11-20 14:41:58 Re: reindexdb: Change quoting rule of --schema?