[BETA] A Z-Wave Mesh Tool [C7 and 2.2.4+ Only]

jlv · December 1, 2020, 9:29pm

I'm wondered if this was because the Iris smart plug I paired earlier didn't join successfully and left a ghost:

But I removed that and I still don't see anything.

eppower · December 1, 2020, 9:44pm

Same behavior for me on a C-7 with 2.2.4.148.

tony.fleisher · December 1, 2020, 9:45pm

Interesting. I haven't seen that one in awhile.
Can you open javascript console and see if there are any errors?

jlv · December 1, 2020, 9:54pm

That helped me find the issue! I access my hub via a name (http://hubitat-hub) and thus my login cookie is stored under the name. The network trace on the console showed me it trying to get to http://10.0.0.67/zwaveInfo and being redirected to the login page. Once I logged in at the IP address, the mesh details page worked on a reload.

tony.fleisher · December 1, 2020, 9:57pm

Great. Thanks for the details (and glad you have a work-around). I will see if i can get this fixed in the next update.

@eppower Thanks for the report. Can you see if the workaround detailed by @jlv here fixes the issue for you also?

eppower · December 1, 2020, 9:57pm

Exact same issue for me, and same solution

tony.fleisher · December 4, 2020, 4:18am

v0.1.16-beta

Now available for install from HPM (please use HPM matchup so you can stay up to date)
Add advanced options to App Settings
- deviceLinks?: Device names will link to device edit page on hub
- Host override: Allow you to change hostname in links (should fix issue reported by @jlv and @eppower)
Refactor Neighbors list and add NeighborsOf (@dennypage: color will be added in a future release)
Instead of coloring red, strikeout Status if not OK and device is not a repeater (idea from comment by @danabw in another thread)

danabw · December 4, 2020, 4:20am

I deny everything. Unless there is a residual. In which case, yes, it was me.

eppower · December 4, 2020, 4:40am

Works well, thanks!

jameslslate · December 4, 2020, 9:58pm

Thanks for this, @tony.fleisher. Works great!

672southmain · December 4, 2020, 10:21pm

Tony (@tony.fleisher), great app.

Simply FYI, I added a device today (GE/Jasco 26931 Smart Motion Switch, running Jason Bottjen’s (@JasonJoel’s) Component Motion Switch driver). The mesh is in the process of settling and re-routing, everything working perfectly.

Z-Wave Mesh Details shows “Data Error” in RTT StdDev column. Hard to understand how a standard deviation calculation could return a Data Error, thought you might want to see this. Repeatable, only for this device (node 0x36, the last node). No errors in logs. 2.2.4.156 on C-7.

Z-Wave Mesh Details screenshot

Z-Wave Details Screenshot

dennypage · December 4, 2020, 11:03pm

I think it might be coming from the raw data. Check the device in the URL: http://YOUR_HUB_IP/hub/zwaveNodeDetail

The fields are "sumOfTransmissionTimes" and "sumOfTransmissionTimesSquared"

672southmain · December 4, 2020, 11:22pm

Sigh. Before I saw your post, I had rebooted the hub during adding a smart plug for the Christmas tree. This is my first reboot since doing the 2.2.4.156 update when it came out. The error is gone after reboot, perhaps an early present from Santa. I’m still hoping for a pony.

scottgu3 · December 4, 2020, 11:24pm

I've seen that error a number of times on my C7 mesh as I was building it out. I don't think it's indiciative of anything in particular, except perhaps a lack of data.

S

p.s. Good luck with the Pony.

tony.fleisher · December 4, 2020, 11:49pm

Thanks. I was wondering if anyone else would come across this.
This happens when there is something apparently wrong with the raw data used for calculations, such that the StdDev calculation would try to calculate the squareroot of a negative number, which isn't valid. I was going to send some details to @bcopeland about it, but it went away for me after reboot for one of the recent platform updates and I haven't seen it again.

scottgu3 · December 4, 2020, 11:56pm

I had a device that hung onto it for a while, I briefly started panicking, but then I decided that was too much effort and chilled right out.

Seriously though, It did stick around for some number of hours.

To be fair though the device in question is in my Faraday room (utility) which seems to consist of mostly large metal things with a bit of air around them!

S

tony.fleisher · December 5, 2020, 12:02am

Thanks for the feedback. Sorry for causing undue alarm. I think it is caused by some error in updating the statistics (or possibly counters that overflow). I only saw it on devices with very high RTT.
I will come up with a better label in a future release.

scottgu3 · December 5, 2020, 12:05am

Honestly, I just figured it was a divide by zero error in the averaging caused by a lack of data from the device. I figured it would eventually go away. I probably should have looked at the code to see where it might be coming from, but my Groovy Fu is a bit weak!

S

tony.fleisher · December 5, 2020, 12:09am

You were close.
IIRC, it is when the sum of the squares is less than the sums squared

Panda · December 5, 2020, 3:36pm

I have the Data Error on one of my devices if you are still trying to track it down.

Any idea what a negative LWR number means? I assumed this was a signal strength number but I have a lot that are red and am not sure if I should care.

Ok also don’t get why I still have unknowns since the hub must have set a packet by now to all my devices and especially my repeaters. I assumed this would clear quickly after a reboot but they don’t for a lot of mine which I have just accepted.

I rebooted last week to reset all the numbers. Still getting a lot of route changes which was surprising but actually observable with this report!