Report Metrics for the driver

[leakingtapan]

Is your feature request related to a problem? Please describe.
Report Metrics for the driver

Describe the solution you'd like in detail
TBD

Ref

• Volume usage metrics support container-storage-interface/spec#236

[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

[leakingtapan]

/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

[frittentheke]

/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

[zerkms]

/remove-lifecycle stale

[rgarcia89]

Awesome to see that this will be part of milestone 0.6
very looking forward to it

[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

[zerkms]

/remove-lifecycle stale

[AceHack]

What's going on here, how do we get metrics, this seems very incomplete without it.

[konduktor4ik]

Any update?

[AceHack]

Any update, I really hope this is done before 1.19 when CSIMigrationAWS is likely to be enabled. This will break tons of our alarms and auto-resizing code if we can't get metrics.

[Ramneekkhurana]

can we have any updates on this ?

[ayberk]

We're currently not actively working on this but we understand how important it is. I'll see if we can get some resources soon.

@AceHack I believe currently it's planned to be on by default with 1.21 release.

[AceHack]

Any update?

[AceHack]

Great updates, thanks so much for the work on this.

[kovalyukm]

Please, tell in which version of the driver and when volume stats metrics are going to be available? (For now #677 is in master only).

[ayberk]

We'll have it in the next release, so 0.9.0.

[Pluies]

For those wondering (as I did!) what metrics are now available, as I couldn't find any info about them in this project's docs, they are the standard:

kubelet_volume_stats_inodes_used
kubelet_volume_stats_used_bytes
kubelet_volume_stats_available_bytes
kubelet_volume_stats_capacity_bytes
kubelet_volume_stats_inodes
kubelet_volume_stats_inodes_free

And they're exposed on the kubelet metrics endpoint without any extra config. So if you have e.g. Prometheus scraping the kubelet, the metrics should be available already! 👍

Thank you for adding this @AndyXiangLi , it's super helpful 🙏

[augustovictor]

Even though these metrics are now available, are they collected by cloudwatch as well or it'd be necessary to configure prometheus to scrape them, for example?

[ghost]

I'm in a similar situation like @augustovictor - my EBS driver is deployed as an addon via Terraform.
I also have enabled CloudWatch integration (addon) with Container Insights.
Running commands such as kubectl get --raw /api/v1/nodes/(HOST)/proxy/metrics | grep 'kubelet_volume_stats_available_bytes' does indeed show metrics but how do I get these metrics into CloudWatch so I can create alarms?