I think we also need to urgently write docs for vendors... Life is so much better if the device sends us their warning and critical limits!
I'm sure we could do a better job, too, grabbing some averaging data over time or something (or maybe even in our hardcoded device hardware tables) but it's hard to differentiate between "yes, 200rpm is normal" and "no, 200rpm, are you crazy?! it's broken!!1". I have no real idea how to improve our current method ;/
Tom
On 13/12/2017 03:58, Adam Armstrong wrote:
Perhaps we should write some docs on this subject, I think a fair few don't understand how it works (or doesn't sometimes)!
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=11327 On 13 Dec 2017, at 02:50, Jacob Bisby <Jacob.Bisby@uniscope.com.au mailto:Jacob.Bisby@uniscope.com.au> wrote:
Yep implementation is pretty much flawless - my only bugbear is I've had to set custom limits for the fan-speed (1900 low, 6000 high) because the autotuning seems to get caught out by the fans ramping up when the device is polled. Jacob Bisby Network Engineer Unit 3/10 Advantage Way Wangara WA 6065 +61 865 555 850 +61 415 073 237 jacob.bisby@uniscope.com.au www.uniscope.com.au <http://www.uniscope.com.au>