I have read on this Forum that Hubitat constantly "churns" the Z-Wave mesh. That is, it is regularly trying to find a more optimal Z-Wave mesh.
I have not added/deleted/moved/changed a Z-Wave device in a while. Yet every time I look at my Z-Wave mesh it is dramatically different. Sometimes every thing is working great for a week or two, and then I have a problem for a few days. I have 2 devices that seem to be placed such that they readily act as repeaters. One of these devices will be "repeating" 3 other devices one day, and will be repeating 9 other devices the next day, and then 6 other devices the day after. While I appreciate that Hubitat may be trying to optimize my Z-Wave mesh, it never ends the process and settles down to a "good overall" mesh. Sure, I want the mesh to be able to adapt to problems/dead devices. But the amount of change from day to day is extreme and seems to serve no purpose.
That isn't really Hubitat per se; it's just how Z-Wave works (at least ignoring the upcoming Z-Wave Long Range specification). So, the answer is basically "no" (though this is also ignoring the fact that "classic" Z-Wave devices might need a repair to do this, so just not doing needless repairs is one way to avoid this issue; I'd assume yours are doing it on their own, and this is probably not applicable for your devices).
But in general, I don't believe a device won't deviate from its last known working route unless there is a problem. There are people here who can tell you more about the nitty-gritty details than I can. But one way you might have some control over this situation is to strategically place your repeaters in reliable locations, maybe ensuring you have one in good range of your problematic devices as well as the hub itself. The Z-Wave Details page may be able to give you some clues, which it sounds like you've already checked out, though in general. I'd really only look there if you're having problems (which it unfortunately seems like you do on occasion? any details about the specific issues and devices might be helpful if you're interested in troubleshooting that).
So I guess the long answer is "no," too, just that there might be some other things you can do to help.
The 700 series SDK does it’s own thing… All we do is periodically send a Node Neighbor Update Request.. Nightly for mains powered and weekly for battery devices.. Once the SDK gets an updated node neighbor list it can choose to re-route things.. But this will often happen with no interaction at all, if packets get lost through a route it will start changing the routes too..
This can even happen between commands.. The route that is displayed is known as the LWR (Last Working Route).. The SDK actually maintains a list of multiple routes to a device, and the device has multiple routes back to the hub.. The last one used to successfully send a packet is what is displayed.
In a stable network (mesh), I don't see the need to perform a daily Neighbor Update. It is extremely rare for a device to fail, and if I add/move devices, I always perform a Z-Wave Repair.
It occurs to me that your performing a daily Node Neighbor Update Request could be resulting in (aggravating) in the 700 series SDK constantly updating the Route. Would it be difficult or unwise to have an option to halt the daily Node Neighbor Updates? I would be happy to test this kind of option to see what it does to my Routes.
