I own a C5 model and just recently started giving it a new chance.
One of my main issues with it in the past is that it kept freezing up, so few months ago i un-installed everything custom from it and tried again. I have had a lot of help from the comunity, so thought to ask for suggestions as well. I have almost given up on finding the reason, so i am looking for a fire and forget solution that keeps my hib from freezing.
I moved a bunch of essentials to the hub, like heating, so few days ago it froze up, just after i solved the Away mode and heating. But this time it resulted to my heating going to 100% and spending gas for 4 days in a matter of 8 hours. So, i need to solve this.
I have a bunch of NodeMCUs, some arduinos as well, and i could repurpose one of those to test some url of the hub and (somehow) restart it. I manually unplug and replug the power to do this. So an automation would be awesome.
I can use any suggestions at this point before i box my solution too much.
i use hubitat and home assistant for this,
On the hubitat side i use RM to turn on a “reboot hubitat” switch when either zigbee or zwave radioes crash. the “reboot hubitat” switch is also presented to home assistant.
once the “reboot hubitat” switch is on RM tells hubitat to shutdown after 10mins.
On the home assistant side i have http monitors for both my AP and the hubitat diag port.
If the hubitat diag port stops responding for 2 mins and the “reboot hubitat” switch is on or not responding, power off the hubitat power plug for 15 min and power on.
If the hubitat diag port stops responding for 30min and my AP is responding, power off the hubitat power plug for 15 min and power on.
Watchdog Timers and etc. is not a solution for the problem!
This is just a last resort bandaid (but yes, better than nothing).
Instead a real problem better/must be identified and fixed.
I agree, and in my case it was. hubitat firmware was pushed that fixed both the zigbee and zwave radio crash issues. the automations we built to mitigate the issue until it could be fixed. They stayed in place just in case the issues return. It’s been a while since any of them have been triggered and i do get notified when they do.
To me, there is a lot of value to self healing systems. There will be issues and errors, i just dont want to be impacted by them
But, my device is out of support and i don't know how to get those logs that would help me.
The logs i do get just give me the last logged message so i can confirm when it froze and new logs for when i manually restareted it.
So, in essence the logs just confirm that it froze, but i don't get anything similar to a crash log.
The hub is not accessible from the network, so i am not able to get info about why automations failed, the hub is 100% stuck.
So i am in a position to either throw HE in the recycling or automate my manual fix.
Even though you may be past the warranty replacement window, have you asked Bobby Dobrescu (@bobbyD) to take a look at your C-5’s internal hardware logs to see why it is hanging up? It may be a hardware issue, or it may be a bad device attached to your C-5 that is causing the crashing.
Hubitat is good about providing support to customers even out of warranty. I suggest you at least get that information to help you decide on a course of action.
That i have not done, partly from reading other threads about hub getting stuck and i got maybe a wrong impression that out of warranty devices don't get support.
I will ask directly for help and see what hapens, thanks for the nudge.
I can imagine it is busy with the holidays, so i got no other communication than asking for the hub mac etc.
But the hub got stuck today again, and my ping watcher alerted me, thankfully i was at home so i could just restart the hub by unplugging it from the power.
I guess it is time to get serious about a hardware watcher.
