[umbrella] Metrics Stability Migration Implementation

[RainbowMango]

This issue works as an umbrella holding up all related items of implementing KEP "Metrics Stability Migration".

Work Items:

• Wrap buckets functionality and migrate metrics to stability framework
  • Wrap buckets functionality. Add wrapper for prometheus buckets functionality  kubernetes#82583
  • Migrate apiserver. Migrate prometheus bucket functionality to metrics stability framework for apiserver kubernetes#82630
  • Migrate volume. Migrate prometheus bucket functionality to kube-metrics for volume kubernetes#82626
  • Migrate proxy metrics. Migrate prometheus bucket functionality to kube-metrics for proxy metrics kubernetes#82736
  • Migrate proxy winkernel. Migrate prometheus bucket functionality to kube-metrics for winkernel kubernetes#82737
  • Migrate scheduler. Migrate prometheus bucket functionality to metrics stability framework for scheduler kubernetes#82741
  • Migrate stability framework itself. Migrate prometheus bucket functionality for metrics stability framework kubernetes#82745
  • Migrate kubelet. Eliminate direct references to prometheus from kubelet kubernetes#83499
• Wrap labels functionality
  • Wrap labels definition
• Introduce testutil and refactor test cases
  • Introduce testutil. Introduce testutil package to support metrics testing kubernetes#83299
  • scheduler. Refactor scheduler metric test with testutils kubernetes#83611
  • kubelet. Refactor kubelet ut with metrics testutil kubernetes#83664
• Wrap GaugeFunc functionality(hold, rely on Provide a mechanism for custom collectors to use the metrics stability framework kubernetes#83062)
• Remove reference to Prometheus
  • kube-scheduler. Remove direct reference to label kubernetes#83220
  • cloud provider openstack. Remove direct reference to prometheus from openstack kubernetes#83269
  • cloud provider vsphere. Remove direct reference to prometheus.Label kubernetes#83223
  • cloud provider aws. Remove direct reference to prometheus from aws kubernetes#83278
  • kubelet. Remove direct reference to prometheus from kubelet kubernetes#83279
  • intergration test
    • Remove references to prometheus is test/e2e kubernetes#85029 @serathius
    • Remove prometheus references in test/integration kubernetes#84283 @serathius
    • Refactor test/integration/master/ with metric kubernetes#83833 @tanjunchen
• Remove reference to promhttp
  • etcd-version-monitor remove reference to promhttp in etcd-version-monitor kubernetes#84221 @wuyafang
  • kubelet remove reference to promhttp in kubelet kubernetes#84393 @wuyafang
• Reference to github.com/prometheus/client_model/go

/sig instrumentation
/assign

[hwdef]

Can you open an issue contains more questions waiting to be resolved?
If you open an issue just serving for yourself,it is meaningless.

[tanjunchen]

@hwdef @RainbowMango Agree! Accelerate migration ! To hold all the releated works about metric refactor in k/k is Good!

[RainbowMango]

@hwdef @tanjunchen (as well as @SataQiu )
I'm glad that you are interested in the feature. Welcome!

Actually, I'm thinking about how to split items for other contributors who want to join.
Please let me know what kind of task you like. good first issue or something with a little challenge?

I figured out the items list above and put them to different iterations. If you can join in, you can pick some tasks, it helps to avoid duplication.

You can ping me by slack.(@tanjunchen please ping me on slack, I can't find you.)
Or contact me by Wechat, @SataQiu knows my ID. You are all server at BoCloud, right?

[hwdef]

@RainbowMango
Please just add all items what should be modified at once.
We can chocie them by ourselves.

[hwdef]

This has nothing to do with the company.
We should contribute to the community in a better way.
We are in a open source community, do something like an open source enthusiast.

[tanjunchen]

@RainbowMango I will pick up intergration test . Ok?

[RainbowMango]

@RainbowMango I will pick up intergration test . Ok?

Sure.
You can try it, please let me know if you need help.

[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

[serathius]

/remove-lifecycle stale

[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

[palnabarun]

/remove-lifecycle stale

[palnabarun]

Hi @RainbowMango, is this an enhancement? I see that the issue acts as an Umbrella issue for several changes in k/k.

Can we close this if this issue is not an enhancement since k/enhancements repo is used for tracking KEPs and corresponding enhancements issues?

[RainbowMango]

ok.

[palnabarun]

Thank you @RainbowMango. 🙂