PostgresNode::backup uses spread checkpoint?
Hackorum builds and tests every patch posted to the lists, not only commitfest submissions. This is Hackorum's own CI rather than the PostgreSQL project's, and it is still under testing - please report anything that looks wrong.
You can run a PostgreSQL built from this patch straight from Docker, with no checkout and no build:
docker run --rm -p 5432:5432 ghcr.io/hackorum-dev/postgres-patch:t43117psql -h localhost -U postgresBuilt from patchset v1 (message #1), July 27, 2026 at 06:19 PM.
Every patchset is also pushed to a branch of our PostgreSQL fork, so you can check out the same tree CI built. Without a PostgreSQL checkout:
git clone --branch t43117_1 https://github.com/hackorum-dev/postgres.gitIn a checkout you already have, add the fork once:
git remote add hackorum https://github.com/hackorum-dev/postgres.gitthen, for this patchset and every later one:
git fetch hackorum t43117_1 && git checkout t43117_1Patchset v1 (message #1) is on t43117_1
I noticed a few days ago that method backup() in PostgresNode uses
pg_basebackup without specifying a checkpoint mode -- and the default is
a spread checkpoint, which may cause any tests that use that to take
slightly longer than the bare minimum.
I propose to make it use a fast checkpoint, as per the attached.
--
�lvaro Herrera
On 10/20/20 11:01 AM, Alvaro Herrera wrote:
I noticed a few days ago that method backup() in PostgresNode uses
pg_basebackup without specifying a checkpoint mode -- and the default is
a spread checkpoint, which may cause any tests that use that to take
slightly longer than the bare minimum.I propose to make it use a fast checkpoint, as per the attached.
+1.
--
-David
david@pgmasters.net
On Tue, Oct 20, 2020 at 11:13:34AM -0400, David Steele wrote:
On 10/20/20 11:01 AM, Alvaro Herrera wrote:
I noticed a few days ago that method backup() in PostgresNode uses
pg_basebackup without specifying a checkpoint mode -- and the default is
a spread checkpoint, which may cause any tests that use that to take
slightly longer than the bare minimum.I propose to make it use a fast checkpoint, as per the attached.
+1.
+1.
- $self->host, '-p', $self->port, '--no-sync');
+ $self->host, '-p', $self->port, '-cfast', '--no-sync');Some nits: I would recommend to use the long option name, and list
the option name and its value as two separate arguments of the
command.
--
Michael
On Wed, Oct 21, 2020 at 07:55:18AM +0900, Michael Paquier wrote:
Some nits: I would recommend to use the long option name, and list
the option name and its value as two separate arguments of the
command.
For the archives: this got applied as of 831611b.
--
Michael