The current volume usage is only available under the specific condition that kubelet endpoints report the necessary volume metrics. In the case where these metrics are unavailable, the service will default to simply reporting the configured capacity level of the persistent volume claim.
In my use case, this seems to work when using a local disk of the node system, but be unavailable when using a hoster-provided volume.
I wonder if someone is aware of a method to monitor PVC utilziation even when the kubelet, which is apparently a node-based control process, endpoint does not report it.
Still struggling with some volumes I so far can only monitor manually. I’ve somewhere/somehow came across the theory that it might be due to the mode not being ReadWriteMany but am kind of stumped on what to do about them and have been sidelining this as I’d have to deploy an entire sandbox cluster from the vendor with all the costs that incurs which seems way too excessive just to test this one damn thing without breaking anything else, because the vendor-specific storage type is not replicable in a local sandbox, and changing the mode faces that same issue and seems kind of dangerous.
Maybe it’s possible to write a workaround to retrieve this with just a script, but facing some obstacles on that front as well regarding sessions…
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.