Test-infra: Allow people outside of Google to be on test-infra rotation

[spiffxp]

• What are the responsibilities of the test-infra rotation role
• What knowledge or experience is necessary to fulfill that role
• What capabilities or credentials are necessary to fulfill that role

[kargakis]

cc: @leifmadsen @stevekuznetsov

[madhusudancs]

cc @kubernetes/sig-federation-misc

[xiangpengzhao]

+1

[spiffxp]

/lifecycle frozen
The intent is to migrate from a google.com-owned GCP org to one that is owned by the CNCF. At that point we are capable of adding people to support our infra, and we can iterate on what exactly "support" means and what requirements there are.

I've documented as best I can what test-infra we're running where on internal clusters. Next step is to figure out a migration plan. https://github.com/kubernetes/test-infra/issues/5085

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten
/remove-lifecycle stale

[BenTheElder]

/remove-lifecycle rotten

On Fri, May 11, 2018 at 3:36 PM fejta-bot [e-mail] wrote:

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta
https://github.com/fejta.
/lifecycle rotten
/remove-lifecycle stale

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/kubernetes/test-infra/issues/3349#issuecomment-388502332,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AA4Bq3HqBNF45DxiUSGL_FCoUx6AOqrcks5txhKFgaJpZM4OPA3Z
.

[BenTheElder]

/lifecycle frozen
this is pending on non-googlers having access to the infrastructure (pending the projects not being funded by being under the google.com GCP org...), once we have something for that we can sort out the rotation details. until then they won't help much.

FWIW, I'm writing a doc with @amwat proposing how the infra-transition could work if funding etc. is sorted.

[spiffxp]

/lifecycle frozen
This is still relevant even if nothing's happening on it right now. There are folks inside and outside Google making a push on the parts that come before this. If it turns out there's a better repo to track this issue I'll move over there. For now I'd rather this stay frozen.

[spiffxp]

/milestone 2019-goals

[spiffxp]

/lifecycle stale

[spiffxp]

/lifecycle active

[spiffxp]

/lifecycle frozen

[spiffxp]

(that was all me confirming that lifecycle labels get swapped out rather than stacked)

[spiffxp]

/wg k8s-infra
/milestone v1.16

Regardless of where prow.k8s.io lives, we need to be able to create and staff an on-call schedule in a way that can handle community members. At present we use a google-internal tool to generate a rota based on fairness and people's calendars.

@rmmh attempted to rewrite the tool a while back, we could try picking that back up https://github.com/kubernetes/test-infra/pull/4688

We could have humans manually manage recurring events on a shared google calendar (this will be seen as toil)

Possible open source on-call things to examine:

• https://github.com/arachnys/cabot (last commit jan 30)
• https://github.com/openduty/openduty (last commit may 8)
• https://github.com/linkedin/oncall (last commit jun 18)

[cblecker]

The PST uses Opsgenie via the CNCF for their rotation (another option).

We still also need the initial questions answered (what is on call responsible for, etc) and documented.