Mailing List Archive

CI status update: Travis out, GitHub Actions in
Support for Travis has now been dropped, so we are relying only on
GitHub Actions for CI from today.

There are some gaps in the GHA configuration compared to what we had
running in Travis: notably, we're currently only testing on the latest
Ubuntu release, and nothing has been ported back to 2.4.x yet. I will
propose a vote separately for the 2.4.x RTC exception to extend to CI
configuration.

On the positive side, the GHA configuration is a lot more flexible, and
e.g. not triggering CI on docs changes was a simple change. Any help
filling remaining gaps would be very welcome, there is a TODO list in
test/README.ci.

Another significant change is that GHA is configured to require approval
on PRs from non-committers - currently only for the first PR filed, but
from March 19th the ASF default will switch to requiring approval for
every PR filed by a non-contributor. I don't think this is a big deal
but will require committers to actively review and hit "approve" on each
such PR to get it tested.

I'm happy to try to answer any questions about GHA, but I'm still
learning how it works so very much a non-expert here.

Regards, Joe
Re: CI status update: Travis out, GitHub Actions in [ In reply to ]
On 2/15/23 11:43 AM, Joe Orton wrote:
> Support for Travis has now been dropped, so we are relying only on
> GitHub Actions for CI from today.
>
> There are some gaps in the GHA configuration compared to what we had
> running in Travis: notably, we're currently only testing on the latest
> Ubuntu release, and nothing has been ported back to 2.4.x yet. I will
> propose a vote separately for the 2.4.x RTC exception to extend to CI
> configuration.
>
> On the positive side, the GHA configuration is a lot more flexible, and
> e.g. not triggering CI on docs changes was a simple change. Any help
> filling remaining gaps would be very welcome, there is a TODO list in
> test/README.ci.
>
> Another significant change is that GHA is configured to require approval
> on PRs from non-committers - currently only for the first PR filed, but
> from March 19th the ASF default will switch to requiring approval for
> every PR filed by a non-contributor. I don't think this is a big deal
> but will require committers to actively review and hit "approve" on each
> such PR to get it tested.
>
> I'm happy to try to answer any questions about GHA, but I'm still
> learning how it works so very much a non-expert here.
>

Thank you very much for all your all efforts on moving from Travis to GHA.
Very much appreciated.

Regards

RĂ¼diger
Re: CI status update: Travis out, GitHub Actions in [ In reply to ]
> Am 15.02.2023 um 11:51 schrieb Ruediger Pluem <rpluem@apache.org>:
>
>
>
> On 2/15/23 11:43 AM, Joe Orton wrote:
>> Support for Travis has now been dropped, so we are relying only on
>> GitHub Actions for CI from today.
>>
>> There are some gaps in the GHA configuration compared to what we had
>> running in Travis: notably, we're currently only testing on the latest
>> Ubuntu release, and nothing has been ported back to 2.4.x yet. I will
>> propose a vote separately for the 2.4.x RTC exception to extend to CI
>> configuration.
>>
>> On the positive side, the GHA configuration is a lot more flexible, and
>> e.g. not triggering CI on docs changes was a simple change. Any help
>> filling remaining gaps would be very welcome, there is a TODO list in
>> test/README.ci.
>>
>> Another significant change is that GHA is configured to require approval
>> on PRs from non-committers - currently only for the first PR filed, but
>> from March 19th the ASF default will switch to requiring approval for
>> every PR filed by a non-contributor. I don't think this is a big deal
>> but will require committers to actively review and hit "approve" on each
>> such PR to get it tested.
>>
>> I'm happy to try to answer any questions about GHA, but I'm still
>> learning how it works so very much a non-expert here.
>>
>
> Thank you very much for all your all efforts on moving from Travis to GHA.
> Very much appreciated.

Thanks Joe!

>
> Regards
>
> RĂ¼diger