Would have been interesting to see if a 500 series stick would have produced a different result from the 700 series stick you were using (I’m pretty sure the UZB-7 is a 700 series stick).
Given both the stick and Hubitat produced the same result (especially with them both thinking they are receiving responses), makes me think it could have been some sort of 700 series underlying SDK bug, but that would just be wild speculation.
Both are Macs with VirtualBox. The VMs are actually clones, although Simplicity Studio has been separately updated since cloning. Power should be fine for both. I generally use the laptop for all my Z-Wave work, and it's not been a problem before.
While the two devices were about 2' apart, the laptop has the stick on the side and is able to communicate with the hub without routing. The workstation routes because the stick is on the back side as far as the hub is concerned, so it hops through one other node.
Apologies, I didn't mention this. I had also tried an old Aeotec Gen5 Z-stick, which also didn't work. And as Lewis mentions, the UZB-7 worked find in the end after moving it to a different computer.
@bcopeland I think it would be helpful to update this original post with current status - I’m holding out on the purchase of a C7 until things have settled down- but this thread has gotten very long and difficult to understand whether these issues have now been addressed and what’s still problematic vs fixed.
Similar to my situation. Although, I'd more characterize it as about 20-30% of the devices in a "large automation" don't respond as they should nearly every time (they seem to clear up after a refresh and a second pass). Plus, there are times when actions seem delayed more than I would expect.
I'm trying to work on this and see what I can do to sort it out.
Btw, @spam2, do you have a lot of "S2" devices paired in "S2 Authenticated" mode? If not, are they mostly Z-Wave Plus devices or not? Thanks
Same, mine were Aeotec tri-Sensors and they included easily once I found the pin.
Out of my 40 odd z-wave+ devices, only those 3 are S2, the rest are S0 or no security and my mesh works great. Even when we leave home and it turns all the lights off in one go, our C7 has been very reliable.
Unless you have non-plus devices, imo, there is no good reason to avoid the C7.
Coming from a Vera Plus, it has been a really great experience - even the early z-wave teething troubles were easily resolved with the help of the experienced HE folk here.
I am seriously thinking that may be the issue, then. I have enough S2 Authenticated devices to potentially be concerning. I am going to have to make a project of getting them re-paired as Unauthenticated so I can see if that helps.
I would do the same if they were not such a PITA to pair. It took me around 10 minutes or longer trying to get them to be recognized. I had to do several Inclusions and removals but this was on early firmware.
Well. Mostly if you care for your hub to know the current status of the devices (switches/dimmers, etc). With the old GE/Jasco ones, that required "polling", which could tie up the network.
Also, ZW+ has a more powerful signal AND it dynamically adjusts routes--hopefully picking better ones when needed. Old ZW kept the same route, even if it was far from ideal.
That’s an interesting tid bit. Does the polling really slow down the mesh network? I have several of the jasco 3way switches in my mesh and have issues with certain all devices on or off scenes.
Depends on the devices. I was doing a Vera install with regular old Zwave GE/Jasco switches about 20 of them. I had to battery power the Vera and have a 100 foot ethernet cable to take the Vera by each switch to have them include. But that was before network inclusion was implemented. It was a pain but after a Z-wave repair (Vera ran one every night) the system was rock solid.