the wotrkaround i found (not really a workaround .. you have to start over) but you can use the hub. is to backup to 2.2.2 , reboot, do a zwave radio reseet, reboot and start readding all devices..
at least hub works again.
so for the past two days adding zwave devices including s2 devices has been working without a problem. I decided to add another aeotec range extender 7. Brand new from amazon yesterday. I paired it, got the popup where it asks for the pin, i type it in and it sits with a "found new device with ID" and nothing more. I check in the zwave details (where the routes can be found) and i see the device id there but it has a button that says "discover" rather than a name. I exclude and click the button, i get unknown device excluded. I reboot and i attempt to re-pair the device. Same thing as above.
Is there some extra magic i should be doing here?
I'm on my fifth go-round of "start over." I'm waiting for the fix this time. 
2 Likes
Why is it unnecessary for Plus devices?
Because whether you like it or not, if they don't have a good path to the hub they will send out explorer frames and rebuild their route automatically. Non-plus devices can't do that - they will just die and stay dead (which is why repairs can help greatly with those devices).
That said, if you put a new device right in the middle of your mesh, doing a repair can help the existing nodes consider using that new device faster. So that is one place where a repair MIGHT help on plus devices.
But again, if zwave plus devices have working routes to the hub, they will tend to stick to those routes until they don't work. And if/when they don't work, they will update their routes automatically anyway.
And on the hub side it is a whole different story - it is always looking at routes and route optimization. A repair doesn't really do much for the hub side.
1 Like
The number one thing I would recommend is to not pair the RE7 securely. Uncheck all the boxes when you pair it. Security for a repeater is not particularly important. Note that even in insecure mode, it may still take a minute (literally) for the hub to decide that the RE7 is completely set up.
If you are interested in a driver (you don't need one), you can find one here. The driver is useful for conducting power and range tests to confirm placement.
I thought I saw that the process of route optimization in plus will take a few days (?) and that doing a repair is just a good way to optimize the routes quicker. Do a manufacturers implementation of Zwave influence this?
@bcopeland Knows more about it than I do (and will likely point out all the ways I'm oversimplifying the situation, if he has time). But no, I have not found that route optimization takes "days" on zwave plus devices.
I have always heard that repeated, but in many hours looking at sniffer logs and both hub and device route data I have not seen that to be true at all.
But it is a complex topic. So sometimes things can be true in a certain context, and not in others. Or true, but not in a particularly useful or meaningful way.
1 Like
oh boy... I've used the zwave toolkit to provide some insight into my ST Zwave network (hopefully a thing of the past , when he sorts out its Zwave issues) and while it was interesting it really only showed how bad my outlying sensors were.
I also read in here some place that more than two hops is not good, but I think some of my devices have 4 or 5 hops. Thoughts?
This discussion may not belong here, I have two brand new C7's that are just sitting with half of my devices on nonfunctioning hubs... so other than waiting for a fix, which sounds like its looking for beta testers, I just found your comment interesting.
I don't subscribe to that school of thought. 15-20% of my 70+ device mesh are devices at 3 hops, and they work great. I prefer not to have anything at 4 hops, though.
1 Like
There isn’t any specific numbers that are the same for everyone.. It will vary widely based on the devices and wireless propagation properties of the home.
1 Like
I feel quite frustrated with this hardware release. It really feels like alpha release and not properly vetted. Maybe I'm alone in this sentiment but it's quite frustrating. For everyone saying this is the first c7 product and that things need to be ironed out I feel like this was rushed to be the "first to market" instead of something actually functional. I paid money for this, it should work at least as well as my prior C5 and bottom line is that it doesn't. Even zigbee has a harder time including compared to my c5. Just very frustrated with my overall experience thus far and I've wasted far more time than I would have liked up to this point. Going to put this thing back in its box until it can at least can function as well as my preexisting c5. Vent over, thank you.
I got it sorted today. I excluded the range extender. I shutdown the hub, removed power for 30s. Then I moved the range extender to a few feet of the hub. Powered up the hub and the range extender included with s2 without a problem!
Just sharing to benefit the community.
2 Likes
While there are undoubtedly some Z-Wave issues to work out, there were no Zigbee changes on the C-7 compared to the C-5, so any problems there are likely of different origin. Have you verified that your (I assume) two hubs use non-overlapping Zigbee channels and that both are good channels for your area, for example? I think Hubitat is supposed to guess what it thinks is best when the hub is first set up, but I'm not sure about the details of how that is implemented. This is just one possibility (again if you have two hubs, you'll need to consider how you can build a strong mesh on one while still keeping it that way on the other, for example).
You're not alone regarding Z-Wave, but I know they're working to get these issues resolved. Hopefully this will get better soon!
4 Likes
I think it is actually not too bad. I learned that S2 devices should be initially setup close to the hub. non-authenticated devices can be setup anywhere. I just setup 4 zwave devices right now and 4 zigbee without any problem. From my vantage point, as long as you follow the steps, it seems to work. Anything S2 needs to be done close to the hub, other than that there are no real gotchas here.
i have 26 devices on my hub. mostly zwave. i got a bunch of the lowes iris motion detectors which i really like, but other than that this is mostly just zwave. Seems stable, or at least for me over the past few days once i figured out the range extender.
This was required for some secure devices like locks and garage door openers in the past, but I don't think it's generally good advice. Someone correct me if I'm wrong. In particular, 700-series Z-Wave devices must support both S2 and network-wide inclusion, so they should pair fine wherever they are in the mesh via whatever method you choose; I believe these features are or at least were both optional (I know S2 later became mandatory) in 500-series devices but still widely implemented.
1 Like
For the most part you are correct @bertabcd1234 the exception being if there are devices in the path that do not support network wide inclusion. But it doesn’t hurt to join closely. The network will settle eventually.
2 Likes
I've not generally had much luck with door locks unless the hub is within a few feet. Sadly, only one is a newer generation one (and the driver isn't picking things up liked I'd expect with that one).
Door locks are usually a bit more trouble.. One of their security “features” used to be including at low power. Which is where a lot of the advice of pairing close to the hub came from.
2 Likes
When can we expect the zwave issues to be fixed 
What will be included in the 2.2.4 release
