#Troubleshooting Unavailable Thread Device
1 messages · Page 1 of 1 (latest)
Do you have multiple apple border routers perhaps ?
We are experiencing these issues for Matter devices, but I guess the same applies for Homekit as that uses the same thread infrastructure
I use Google Home Hubs, and I do have more than one.... In order to get this device added, I had to unplug the further away one (there's one like 12" away from the light).
Any tips here? I'd rather not have to leave the main Home Hub Max unplugged. Maybe if there's a way to disable Thread on the further away one... Right now I have no Thread devices on that side of the house anyway.
Google home hubs do not have trel
That means that you need a solid thread mesh between them
Otherwise packets going to your hub max get lost
That’s why you had to unplug it in the first place
If you had apple border routers they’d use trel and compensate for the gaps in your thread mesh using WiFi
But in the absence of those you’ll need to unplug it or get you’ll need a bunch of mains powered devices
(I think I was the one who helped you get the device working in the first place?)
Right now even though they have the same name and same credentials you might as well have 2 separate thread meshes, and are randomly sending some packets to the wrong one
(Thread uses normal ipv6, there’s no way to prefer a particular border router and there’s no awareness in thread on the lan side of the border routers about which router is best)
They're maybe 25 feet apart, so I'd expect they can talk... Any idea what the distance is? But the Hub Max is 25-26 feet from the actual LED strip so I'm guessing it's connecting to that one...
I could move one of my mains devices I guess, or just try to find a Thread lightbulb that acts as a router.
Or can I use an ESPHome as a repeater? Is that a thing that exists yet? That would be the easiest.
If turning that max off fixes things then either they can’t or it’s defective 🤷♂️
There’s no esphome otbr yet, you can compile otbr for some esp chips but there’s no streamline way yet.
Tbh I haven't tried turning it off yet. Sometimes the device comes back online, sometimes it doesn't, which definitely lends credence to what you're referring to - picking the wrong route.
I had to turn it off to get it joined initially though, so I can try that.
Depending on how well your network is set up it might take a while to heal after you turn it off. Docker and supervisor in particular may take a while to expire stale routes.
This whole thing just reminds me of the classic XKCD comic. This experience is not better than what already exists haha.