I got my hub that crashed yesterday. A rule automatically rebooted it when the Zigbee radio went offline, but that did not bring it back. It did not reboot again presumably because the zigbee radio reported nothing, not online, not offline. That is the second time this happens in a few weeks, and last time, it took 3 reboots, and inbetween, the radio had a « nil » Extended pan ID until the third reboot.
This morning, a second manual reboot this morning brought it back, but in the logs, I saw « Queue full » errors.
What is causing this, why does it make my hub crash so hard that a single reboot does not bring it back ?
Reboots don't affect the radio. It pretty much just clears the queue, for the purposes of this discussion. If the radio, out on the Zigbee SOC is 'jammed up' then a reboot does nothing to clear it. Only a power cycle will affect the radio. Repeated reboots will presumably help the SOC help itself by giving it a few seconds where nothing is trying to use it.
If you ruled out that you don't have any misbehaving devices, then you may be dealing with a hardware malfunction, which would be covered by the hub's warranty.
If your hub is under warranty or if you have Hub Protect, you may want to consider creating a warranty claim by visiting support.hubitat.com.
It has been more than 90 days though. It was behaving just fine before. Problems started when the firmwares for the C8 started coming out.
I am not sure how to "rule out" misbehaving devices though, how do I identify one ? Is there any way to know what is the queue, what device or app may be causing it ? because there has been instances of "queue full", but the logs and device/app stats don't seem out of the ordinary.
