WSL2: Prometheus Daemon Set "node-exporter" failed to run

[anthopark]

Hi

I'm running 1.23.3 on a WSL2 Ubuntu 20.04 (Windows build: 19042).

When I enabled the Prometheus add-on, I noticed the pod for node-exporter is not starting. (CreateContainerError)

The container image is quay.io/prometheus/node-exporter:v1.1.2

I see the message on the dashboard

failed to generate container "<random pod name>" spec: failed to generate spec: path "/sys" is mounted on "/sys" but it is not a shared or slave mount

Because node exporter not working, I'm getting the Bad Gateway error on Grafana as well.

How should I fix this?

Thank you

[ktsakalozos]

Hi @anthopark

Could you try this prometheus-community/helm-charts#467 (comment) ?

[anthopark]

Hi @ktsakalozos,

I was able to get it working by removing the mountPropagation under the volume mount sys while editing the daemon set (microk8s kubectl edit ds node-exporter -n monitoring).

But I'm not sure what effect this solution will have. and I still don't get any metrics from Grafana. Even the node/cluster is not detected on Grafana.

I didn't have this issue when running with Multipass, but I encounter this in WSL2.

Lastly, The literal command from the thread you shared does not work because the index for the volume mount is different for the Microk8s version of ds/node-exporter, which has only 2 volume mounts (hence 0, 1).

[billyors]

Issue persists. I was able to use the workaround from @anthopark but still experiencing the Bad Gateway error with the grafana datasource (prometheus).

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.