From: | Michael Paquier <michael(at)paquier(dot)xyz> |
---|---|
To: | David Steele <david(at)pgmasters(dot)net> |
Cc: | Andrew Dunstan <andrew(dot)dunstan(at)2ndquadrant(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)com>, Michael Paquier <michael(dot)paquier(at)gmail(dot)com>, Postgres hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Re: Rewriting the test of pg_upgrade as a TAP test - take two |
Date: | 2018-03-22 01:39:57 |
Message-ID: | 20180322013957.GD2490@paquier.xyz |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Wed, Mar 21, 2018 at 10:51:04AM -0400, David Steele wrote:
> On 3/6/18 4:12 PM, Tom Lane wrote:
> It seems the consensus is that we'll need a build farm update before we
> can move forward with the patch and that we don't need to wait long for
> people to upgrade.
>
> Andrew, do you have a date for the next release?
Even if a lighter version of this patch stripped from multi-version
support (Who uses that actually?) knowing that the buildfarm code has
its own set of modules to do the same may be a good step forward as per
the feedback of this thread people would like to unify the existing
scripts.
We could also try to figure later on how to merge the buildfarm code
into upstream and if it actually makes sense or not, which is something
I am not completely convinced is worth it. This would remove the need
of any complicated refactoring in PostgresNode.pm to handle binary paths
for a single node, at the cost of of course removing cross-upgrade
checks from the tree.
Still I definitely agree that a release of the buildfarm client should
happen first. 9 days before the end of the commit, the window may be at
the end too short for v11.
--
Michael
From | Date | Subject | |
---|---|---|---|
Next Message | Michael Paquier | 2018-03-22 01:42:20 | Re: [HACKERS] taking stdbool.h into use |
Previous Message | Michael Paquier | 2018-03-22 01:33:07 | Re: [PoC PATCH] Parallel dump to /dev/null |