PerfDisk measures the disk_busy metric. The check can be restricted to a single aggregate with the switch ‑‑raid_group.

We are adding a new switch that allows for a different way of monitoring: ‑‑average

Just like with PerfCpu, using the ‑‑average switch compares the average value of all instances, i.e. disks, to the thresholds. In combination with ‑‑raid_group an average value of disk_busy can now be measured and monitored for all disks of an aggregate.

This feature is currently available as testing release 3.5.0_10.

We are also planing to develop an admin-friendly extension: PerfAggr.