empty cloak
Hi All,
As you can see in image1, the value of aggr "data_ssd_n06" is around 75% in AIQUM .
From the harvest metric, I am trying to get the same value by using "aggr_disk_busy"
And from the image 2 shows that the value after average by aggregate's value is only 40-50,
Why is there a big gap between these two value?
Or do I have some wrong logic when calculating aggregate utilization?
amber tiger
Can you share what version of Harvest you are using ? Also, I am not much familiar with this AiQUM page which you are referring and what metric they are using to show this panel, can you give more detail ?
Also, does SM shows aggr disk utilization panel? If yes, then we can compare directly with ONTAP itself.
empty cloak
@amber tiger They are using 24.05
From the AIQUM, you can select the aggregate and switch to performance view, and you can see the utilization chart.
With the harvest,
they are using avg by (node, aggr) ( aggr_disk_busy{aggr="", cluster=""})
No, we don't have aggr disk utilization in System manager.
amber tiger
Hi @empty cloak, We're not 100% clear on how AIQ.UM is doing their calculation, but from what we can tell, it seems like AIQ.UM is using max(disk_busy) for the utilization graph.
The Harvest Aggregate dashboard utilization panel uses avg(disk_busy).
Harvest also exports an aggr_disk_max_busy metric used in the "Top 5 Aggregates by Max Disk Utilization" panel in Disk dashboard as shown below.
Does that panel more closely match what AIQ.UM is showing you?
