There seems to be a setting for .asf.yaml that lets you excempt a team
from branch protection rules (link
<https://github.com/apache/infrastructure-asfyaml/blob/main/README.md#rulesets>)
github:
rulesets:
- name: "Branch Protection"
...
bypass_teams:
- "release-managers"
So if we could have a team defined solr-rm and ask INFRA to add the RM
to that team before a release, and remove after, then we'd have both...
Jan
> 22. okt. 2026 kl. 21:55 skrev Houston Putman <[email protected]>:
>
> We need to push to main for releases though. Do we need to change that to
> PRs? That would be slightly annoying. Other than releases, I like the idea.
>
> - Houston
>
> On Thu, Oct 22, 2026 at 12:50 PM Gus Heck <[email protected]> wrote:
>
>> +1 to protect main only.
>>
>> On Thu, Oct 22, 2026 at 4:35 AM Jalaz Kumar <[email protected]>
>> wrote:
>>
>>> +1 to protect main from any accidental push. PRs raised on github should
>> be
>>> the only way we make changes in main.
>>>
>>> [Q] what's next here? Who owns these setting changes in the repository?
>> The
>>> PMC chair will do or some one from the PMC will pitch & carry this out.
>>> Since the thread was out there for a week so, I thought of asking it
>> here,
>>> apology if this is already being acted upon by someone.
>>>
>>> On Thu, Oct 15, 2026 at 3:26 AM David Smiley <[email protected]> wrote:
>>>
>>>> +1 to protect main. This has happened before and I want to but in a
>>>> control to stop it.
>>>>
>>>> -1 to the other branches that Rahul listed for the reasons he gave.
>>>>
>>>> On 2026/10/14 18:45:25 Rahul Goswami wrote:
>>>>> +1 to protection on main, limited to PR-only merges.
>>>>> I'd leave out the other branches (branch_9x, branch_10x, etc) from
>> this
>>>>> policy though since we often backport. Creating/reviewing/merging a
>> PR
>>>> for
>>>>> each backport would be cumbersome. Also, I know David's backporting
>>> skill
>>>>> takes care of the backports with the labels, so not sure how that'd
>>> play
>>>>> with branch protection.
>>>>>
>>>>> - Rahul
>>>>>
>>>>> On Wed, Oct 14, 2026 at 2:26 PM Andrzej Białecki <[email protected]>
>>> wrote:
>>>>>
>>>>>> Hi all,
>>>>>>
>>>>>> It’s easy to make some missteps and accidentally push stuff that
>> you
>>>> don’t
>>>>>> intend to branches that you didn’t intent to push to ;)
>>>>>>
>>>>>> We already forbid force push on all branches (at least I hope
>> so?). I
>>>>>> propose to put also a protection lock on all branches (maybe except
>>> the
>>>>>> active release branches to make the RM’s job somewhat easier). In
>>>> practical
>>>>>> terms this means that any changes to these branches MUST go through
>>> PR
>>>>>> merges, and direct commits to branches are rejected.
>>>>>>
>>>>>> Cheers,
>>>>>> Andrzej
>>>>>>
>> ---------------------------------------------------------------------
>>>>>> To unsubscribe, e-mail: [email protected]
>>>>>> For additional commands, e-mail: [email protected]
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: [email protected]
>>>> For additional commands, e-mail: [email protected]
>>>>
>>>>
>>>
>>> --
>>> Regards
>>> Jalaz Kumar
>>>
>>
>>
>> --
>> http://www.needhamsoftware.com (work)
>> https://a.co/d/b2sZLD9 (my fantasy fiction book)
>>
