Hi,

Sorry for not replying to this thread... I was working on
post release tasks. It's not completed yet but almost
done. I want to get consensus about this to complete the
last post release task, "Remove old artifacts". I want to
remove unmaintained artifacts from
https://dist.apache.org/repos/dist/release/arrow/ .

> That sounds ok to me, but is there a more precise criterion than "we
> find a problem"?
> 
> In the past, we have from time to time done maintenance releases based
> on annoying bugs/regressions. But not always.

How about the followings?

- Regression (patch release)
- Security vulnerability (patch release or minor release)
- New platform support with keeping backward compatibility (minor release)
  e.g.:
  - https://issues.apache.org/jira/browse/ARROW-18255
    [C++] Support armv6
  - https://issues.apache.org/jira/browse/ARROW-17487
    [Python][Packaging] 3.11 wheels

> I'm afraid "green nightly CI" is more of an ideal than a reality given
> the breadth and complexity of our fleet of CI jobs.

I think so too. So I think that we need to improve our
nightly CI for maintaining multiple major releases. For
example, we may want to add "allow-failures" feature to our
nightly CI but it may hide our real problems...

> there are also regularly regressions due to changes in
> third-party packages.

Right. So I think that maintenance branches should accept
patches for CI that doesn't change library behavior
too. They contain CI configuration changes, test changes and
so on.
e.g.:
- https://issues.apache.org/jira/browse/ARROW-18299
  [CI][GLib][macOS] Failed to install dependencies by Homebrew
- https://issues.apache.org/jira/browse/ARROW-18285
  [R] Fix for failing test after lubridate 1.9 release


Thanks,
-- 
kou

In <0f814dd3-2462-3fa9-e34b-38fe43140...@python.org>
  "Re: [DISCUSS] Maintenance policy" on Wed, 19 Oct 2022 13:41:42 +0200,
  Antoine Pitrou <anto...@python.org> wrote:

> 
> Hi Kou,
> 
> Le 19/10/2022 à 06:29, Sutou Kouhei a écrit :
>> My proposal: We maintain the last major release:
>> * We maintain 9.Y.Z when the latest major release is 9.0.0
>> * We may release 9.Y.Z when we find a problem such as a
>>    security vulnerability in 9.Y.Z
>> * We drop support for 9.Y.Z when we release 10.0.0
> 
> That sounds ok to me, but is there a more precise criterion than "we
> find a problem"?
> 
> In the past, we have from time to time done maintenance releases based
> on annoying bugs/regressions. But not always.
> 
>> I think that we can maintain multiple major releases with
>> not high release cost by implementing the followings:
>> * Green nightly CI
>> * Nightly CI for all maintained branches (maint-X.Y.Z)
>>    * We need to reduce the time taken to CI
>> * ...
> 
> I'm afraid "green nightly CI" is more of an ideal than a reality given
> the breadth and complexity of our fleet of CI jobs.  We still seem to
> have stability problems in some areas (perhaps Acero?) but there are
> also regularly regressions due to changes in third-party packages.
> 
> Regards
> 
> Antoine.

Reply via email to