> On Sep 10, 2026, at 4:12 AM, sebb <[email protected]> wrote:
>
> On Wed, 9 Sept 2026 at 17:09, Rich Bowen <[email protected]> wrote:
>>
>>
>>
>>> On Sep 5, 2026, at 3:51 AM, sebb <[email protected]> wrote:
>>>
>>> On Fri, 4 Sept 2026 at 22:57, Dave Fisher <[email protected]> wrote:
>>>>
>>>> Whatever you guys choose to do just make sure that the P6 work is done to
>>>> make sure it is configured by puppet. If you don’t then you’ll have to
>>>> redo your work if the machine is ever rebuilt by Infra.
>>
>>
>> Ok, I think that what needs to happen is:
>>
>>
>> 1) Check out https://github.com/apache/comdev-metrics somewhere on
>> projects-vm3
>> 2) Put a valid GitHub API key in .secrets in that directory. (I have no idea
>> how to do this securely an in infra-approved ways, and could use schooling
>> here.)
>> 3) Add cronjob to run `uv run asfmetrics` out of that directory every week
>
> Rather than run the job on the VM, could it be run as a GH action?
> In which case, does the action not have access to sufficient karma to
> run the queries ?
>
> That would avoid the need to manage a GH API key.
>
> Or is there some reason why that could not work?
>
> I'm not saying it should not be done on the VM, but it might be easier
> to manage as a GHA.
I know very little about GHAs, but that’s an interesting idea. The queries are
against three data sources — Github repos, the lists.apache.org
<http://lists.apache.org/> API, and (for a handful of projects) svn. It’s all
in Python.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
