CPU usage suddenly went from averaging about 4% to about 35% last night around 8:30pm.
I am trying to see what might be causing this, but the logs show nothing useful (no warnings or errors, log flow looks normal). Automations are running. The Zigbee logs page stays blank (connecting to the web socket using wscat also returns nothing).
The hub has nearly 24 days of uptime so the app and device stats, which are cumulative, are not useful to figure out what is going wrong.
Is there any way to reset those stats without rebooting the hub ?
Upon further investigation, it appears the CPU % shot up at the very moment the zigbee log socket went dead. Bug?
I tried comparing app and device stats by pasting them in excel but it's super tedious (a reset button would be great!) and I didn't find anything interesting. I also tried disabling almost everything (apps & devices), including InfluxDB telemetry, to see if anything made a difference - no difference. The "CPU avg" reported by http://hubitat.local/hub/advanced/freeOSMemoryHistory remains unchanged at around 1.3 (it was 0.2-0.5 prior to this event).
Thanks, however, I am chasing a sudden change. Those are averages since last boot, so after 25 days, the average changes so slowly that you couldn’t detect a sudden change in cpu this way.
When I have seen something like this it was because something got in a bad state and created a race condition. Only solution was to reboot the hub and fix what triggered the issue. Did you change anything recently, add something that talks to maker API frequently?
