[project-monitoring] Summary dashboard has no metrics for Disk I/O

[eumel8]

Describe the bug
On all Project Monitoring installations the graph for Disk I/O are empty. The underlying dashboard Rancher/Pod has also no metrics. Prometheus has some metrics for container_fs_writes_bytes_total and container_fs_read_bytes_totak, but it seems there are some changes , linked with this pr, also mentioned in this OpenShift Issue

To Reproduce
I couldn't full reproduce the issue and what caused the empty metrics exactly. Maybe it needs a device-selector, optional a newer version from upstream (prometheus-federator 0.2.1)

Result
empty metrics

Expected Result
metrics for disc io on all Kubernetes and Rancher dashboard

SURE-6055

[MKlimuszka]

We can check to see if there is an upstream issue and link it, but beyond that it is out of our control.

[bashofmann]

I think this may be fixed with rancher-monitoring 102.0.0+up40.1.2. I tested it with RKE1 v1.24.10.

@eumel8 Can you check if this is fixed with new monitoring/prometheus-federator version for you as well?

[aiyengar2]

Probably a dup of rancher/rancher#38934, which was ported into the latest rancher-monitorng chart (which was brought into prometheus-federator as well).

@eumel8 feel free to open up a new issue if you are still seeing this, closing this out since it is probably fixed

[eumel8]

seems fixed on the updated 102.0.0+up40.1.2 as of today on our clusters. thx, @aiyengar2