Ok. It's done. I went through hell and back getting it set up because my ST hub wasn't in a good way. Why was I doing this again? It's been so traumatic I've forgotten.
I tried talking to it. "Hey, 'lil peanut. Have I got a hook up for you " But peanut wasn't havin' it. Tried two. No joy.
This leads me to another question. Having that ST hub actually turned on, it caused issues with my HE hub because zigbee is on 20 and I see that there's no way to change it. How much hell will I cause by changing the channels on my HE? And what channels are recommended?
Rebuilding the mesh is always fun... Channel recommendations depend on what devices you have, some can only operate on a limited set of channels. The other part to consider is overlap with 2.4G WiFi channels in use.
To be honest, I don't know if it's worth it yet to make the change. I installed echo speaks on ST and noticed the amount of time it takes to talk. I unplugged my hub again and repaired my zwave. For whatever reason, that was going crazy.
It is a major undertaking if you have a lot of different devices, I did it a few weeks ago with 60+ zigbee devices, my mesh wasn't back to normal for a few days. Now it's rock solid and I'm happy I did it since all my devices "like" my current channel better.
For the zwave part I really couldn't say, where I live there really are none to be had, or I'd have to use US ones where the frequency would be illegal.
All of my devices have been happy on 15, 19 (no one on channel 6 by me), and 25. I have Iris V2 motion and contact; Sylvania bulbs, button controllers, motion sensors; GE switches and dimmers; Samsung 2018 & previous gen plugs and contact, and 2018 water sensors; Hue dimmers. The only devices I have to reconnect when I change channels are the Hue Dimmers.
Just a data point. Since removing all but one Peanut plug (another Samsung is in the mail), my getChildAndRouteInfo table seems to be much more populated than previously.
More specifically, it seems to have more entries in both the Neighbor table and the Route table.
Sadly, this is a subjective evaluation, as I have no saved history of the previous results.
I have so many zigbee devices that my tables were always populated. I have noticed that they now are stable, where before they were changing every minute or two. Also, my devices have been rock solid and I can turn off the entire downstairs with one group device without any devices getting missed. Never had that before.
@april.brandt I'm glad you started this thread. I've been quietly watching the rise in peanut allergies too. Maybe like most allergens, it's a buildup over time?
Seriously though, it seems the same rules apply as always.
Not every device is going to work perfectly
Not every device works well together
Inexpensive devices are usually inexpensive for a reason and may be less globally compatible.
I joined two peanuts to @adamkempenich lightify app. They showed up as unknown and I didn't see them right away. I'm going to troubleshoot them this morning and see if I can get some use out of them this way. Looks promising. Adam called it an ol peanut. I called it 'lil peanut. Was there a difference? No one will ever know.
[EDIT] @adamkempenich I tried to pair the device again. It sees it, but lightify doesn't like it. Won't let me interact with it, but it's showing up in HE. I think I'll try to tinker with it.
Oh, but that might indicate that your devices are NOT routing through repeaters.
I understand that end devices do not show up in that list when they are routing via repeaters.
So to be clear, both the Neighbor Table, (which appears to be powered routing devices directly connected to the Hub) is more populous, and the Route Table also indicates more devices connected via the devices now showing in the Neighbor Table.
This remains purely subjective, but in all honesty, previously this screen did not provide me a lot of value. However, I'm now able to see a much larger portion of my Zigbee network displayed.
I don't know if that has to do with both how the Peanuts perform as repeaters, and how they report the routed devices to Hubitat, but this configuration seems superior to me. That is the Samsung ZIgbee outlets seem like they are repeating as one would expect, and the paths displayed seem to make a degree of sense from a Line of Sight and relative location/type of devices connecting.
NOTE: since i think this is relevant, I'm reposting here my experience with hue motion sensors and their effect on my mesh. It was originally posted in another thread
Okay, so without this ghost feature I would have never discovered- That my hue motion sensors, I love so much, are really messing up my mesh.
Constantly having ghost devices be found by the HE feature for ghost zigbee devices. And was finally able to track down to Hue motion sensors. Similar to the great peanut panic, these hue motion sensors are jumping around like crazy and constantly switching their 16 bit address, and those ghosts are found by Mike's feature(i haven't upgraded yet to remove feature)
About 2 weeks ago I added the 16bit address to each of my 4 hue motion sensor's names. ALL except 1 hue motion sensor have changed their 16 bit address, and therein is the cause of my ghosts.
When I hit "get info" on my ghost,I see this below.
For the simple minded among us (me)... can someone point out where I can look at this routing table and see if the many peanuts I have are changing routes frequently as mentioned?
TL;DR: Best to think of that page not as a routing table, but as the status of the 'first hop routers' that your hub will use when talking to devices in your mesh. Some folks have seen Peanut plugs frequently lose/regain their routing neighbor status.
Unlike Z-Wave, where mesh topology is relatively static until there is an indication of a failed message delivery (in which case previously working alternate routes are tried, and failing that, new route discovery is initiated), in a Zigbee mesh the routing neighbor nodes are constantly communicating their status to each other approximately four times a minute, even when no other traffic is going on. That's how it can 'self heal'.
Each Zigbee mains powered routing device (including the hub) tracks up to 16 in-range routers-- shown in the "Neighbor Table Entry' section; each expects to receive status information at periodic intervals that indicates how well it is being heard on the link by means of a 'outCost' number provided by the neighbor (zero if no info has been provided during the expected interval, or a number ranging from 1-7 with 1 being the best measure).
Likewise it transmits an 'inCost' figure to its neighbor link, telling that neighbor how well it is capable of 'hearing' that neighbor. That way routes can be chosen that are good in both directions (chosen so they are good from the point of view of both the sender and receiver). The routing strategy picks routes by these in/out cost figures (rather than LQI, which is a figure of merit for one side of the link only-- the hub's).
The elapsed time intervals between reports are tracked by the 'age' counters; a few intervals are allowed for synchronization of the exchanges. So by refreshing this page you will see the age counters increment (from 3 to 6) depending on when the last link status was received. If the age counter exceeds 6 intervals with no new status, the link is considered stale and can be replaced by another one. So as conditions change (nodes fail or RF environment degrades), neighbors can be evicted from the table and be replaced by other potentially viable links.
With a lot of neighbors in range of each other (not a great scenario for Zigbee, where you actually can have too many repeaters) this table can always be in flux. With a stable network these neighbors may change, but not very often. I haven't used Peanut plugs, but folks that have had issues with them have seen them frequently bouncing in and out of the neighbor table.
I still have and use 10 peanut plugs for a year and half with zero zigbee mesh issues, and live by the if it's not broke don't fix it rule. But no sense in rehashing that debate.
I just counted... I have 22 Peanuts... ugh... and more in a box. I've looked at the routing table a bit and still trying to figure it out... (figure out if I have a problem). I know I suffer from the "hub slowdown" problem and some have linked that to peanut issues...
To complicate matters, Peanuts are variably chatty even with Almond routers. I have two Peanuts (identical firmware) paired to an Almond 3. One of the floods the Almond with power reports; the other is virtually silent. Each of them is connected to the same load (the power-supply to a Hubitat).
" But peanut wasn't havin' it. Tried two. No joy. 
