Drive monitoring issue

[mikeyo]

Whenever a drive is spun down, pulse spins up the drive to grab SMART. This is not good for drive health. Can this behaviour be modified so that SMART values are only polled when the drive is active?

[rcourtman]

This is a feature request/design consideration rather than a bug.

Currently, Pulse polls SMART data at the same interval as other metrics. Spinning up a spun-down drive to check SMART is a known downside of any monitoring tool that collects disk health data.

Options to consider:

1. Separate SMART polling interval - Allow configuring SMART checks less frequently (e.g., once per hour instead of every 30s)
2. Standby detection - Skip SMART checks for drives in standby mode (requires additional detection logic)
3. Manual SMART trigger - Only collect SMART on-demand via the UI

Which approach would work best for your use case? I'''m inclined toward option 1 (separate interval) as the simplest solution.

[mikeyo]

This is a feature request/design consideration rather than a bug.

Currently, Pulse polls SMART data at the same interval as other metrics. Spinning up a spun-down drive to check SMART is a known downside of any monitoring tool that collects disk health data.

Options to consider:

1. Separate SMART polling interval - Allow configuring SMART checks less frequently (e.g., once per hour instead of every 30s)
2. Standby detection - Skip SMART checks for drives in standby mode (requires additional detection logic)
3. Manual SMART trigger - Only collect SMART on-demand via the UI

Which approach would work best for your use case? I'''m inclined toward option 1 (separate interval) as the simplest solution.

Personally, I would prefer option 2. That way, i can set and forget and not have to worry about anything. A quick hdparm check to see if drive is active or standby should suffice, no? Thanks!