How to monitor disk usage?

[cocowalla]

OpenTelemetry is working great for process and runtime metrics, but I'd also like to report on remaining disk space, so I can configure alerts when it's running low.

Is there an instrumentation package available to do this?

[cijothomas]

Is there an instrumentation package available to do this?

Nothing exists in OpenTelemetry repos for this today.

[cocowalla]

Are host/hardware metrics on the road map?

In the meantime, is it just a case of me creating my own Gauge to monitor disk usage, or is there another approach I should use? (I'm new to OpenTel, but liking what I'm seeing :)

[utpilla]

Are host/hardware metrics on the road map?

Not anytime soon. Semantic Conventions has an Experimental spec for it: https://github.com/open-telemetry/semantic-conventions/blob/main/specification/metrics/semantic_conventions/hardware-metrics.md#hwphysical_disk--physical-disk-metrics. Things might change as they make progress towards stabilizing it.

In the meantime, is it just a case of me creating my own Gauge to monitor disk usage, or is there another approach I should use
Yes, you could create your own instrument. Please keep in mind the instrument selection criteria specified in the spec when choosing the right instrument for your requirement: ObservableGauge vs ObservableUpDownCounter.