I'm currently using the netapp probe and noticed the averaged latency isn't accurate.
Is anybody using the ontap api to get the latency values as they apparently aren't available in SNMP? Running ontap 8.
For what it's worth, I'm on 1.22 and using SNMP and I'm getting negative values for latency.. Nice huh? I've opened a case with Nimsoft for that issue.
I'm looking into using logmon and the NetApp Powershell Tookit. That should give me accurate values. I just need to set it up.
I have seen other wild values from the storage probes too. IOP values in the vegative billions or possitive far beyond any sane value. Looks like it make be a number overflowing. I am not sure if the probe is at fault of if the API occasionaly throws a crazy value and the probe just accepts it.
I have written a script I run from the nas sceduler that looks for these extreme values and replaces them with a sane average (so as not to distort the stats too much). If you are interested I will tidy it up a bit and post it here.
Retrieving data ...