Prepare requested resources visualizations for Kubernetes Observability

[eedugon]

On Kubernetes is very common to take a look at the overall resources requested / reserved (not in use) at different levels:

• Cluster wide
• Per node.

When doing kubectl describe node we can clearly see something like:

  Resource                   Requests      Limits
  --------                   --------      ------
  cpu                        1073m (27%)   2210m (56%)
  memory                     6924Mi (52%)  7324Mi (55%)
  ephemeral-storage          0 (0%)        0 (0%)
  hugepages-2Mi              0 (0%)        0 (0%)
  attachable-volumes-gce-pd  0             0

Today the majority of dashboards for Kubernetes monitoring allow the user to visualize quickly the overall (or per node) amount of CPU and memory requested.

In our current dashboards we focus on resource in use, which is nice but we don't have anything about requested resources or limits (maybe limits is not that interesting).

Taking a look at the documented metrics from our Metricbeat module and doing some tests I guess that we have defined the metrics:

• kubernetes.container.cpu.request.cores
• kubernetes.container.cpu.request.nanocores

But I believe (i could be wrong) we are not populating them.

So if we want to achieve this we probably need to solve both parts (receiving the metrics + preparing the UI at some level (dashboards, observability app, ...)

[sorantis]

cc @ChrsMark

[elasticmachine]

Pinging @elastic/integrations (Team:Integrations)

[MichaelKatsoulis]

Taking a look at the documented metrics from our Metricbeat module and doing some tests I guess that we have defined the metrics:

• kubernetes.container.cpu.request.cores
• kubernetes.container.cpu.request.nanocores

But I believe (i could be wrong) we are not populating them.

@eedugon you are right that we are collecting these metrics but the reason you don't see them populated is probably the version of kube-state-metrics. Since v2.0.0 they deprecated those fields (they are replaced with new. Thy used to co-exist for some versions). breaking changes
There is an issue elastic/beats#21520 to support these changes and we will focus on that in the near future.
After the issue is fixed we could also add a visualisation

[eedugon]

Awesome! Great analysis, thanks! El El jue, 12 ago 2021 a las 15:43, Michael Katsoulis < ***@***.***> escribió:

…

Taking a look at the documented metrics from our Metricbeat module and doing some tests I guess that we have defined the metrics: - kubernetes.container.cpu.request.cores - kubernetes.container.cpu.request.nanocores But I believe (i could be wrong) we are not populating them. @eedugon <https://github.com/eedugon> you are right that we are collecting these metrics but the reason you don't see them populated is probably the version of kube-state-metrics. Since v2.0.0 they deprecated those fields (they are replaced with new. Thy used to co-exist for some versions). breaking changes <https://github.com/kubernetes/kube-state-metrics/blob/master/CHANGELOG.md#v200-alpha--2020-09-16> There is an issue elastic/beats#21520 <elastic/beats#21520> to support these changes and we will focus on that in the near future. After the issue is fixed we could also add a visualisation — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1014 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGBFXJP6MYDN3FQYHLYRWQLT4PFZXANCNFSM45JNPYRQ> .

[MichaelKatsoulis]

Awesome! Great analysis, thanks! El El jue, 12 ago 2021 a las 15:43, Michael Katsoulis < @.***> escribió:
…

In matter of fact we are already working on elastic/beats#27552 to support the newer versions of kube-state-metrics and populate the fields.

[ChrsMark]

We are in an effort to improve/optimise our dashboards at elastic/beats#31021. @MichaelKatsoulis maybe these metrics are some to consider too?

[MichaelKatsoulis]

Yes indeed. These are useful metrics we could visualise.

[botelastic]

Hi! We just realized that we haven't looked into this issue in a while. We're sorry! We're labeling this issue as Stale to make it hit our filters and make sure we get back to it as soon as possible. In the meantime, it'd be extremely helpful if you could take a look at it as well and confirm its relevance. A simple comment with a nice emoji will be enough :+1. Thank you for your contribution!