I hope there is someone in the beta who has Z-Wave Classic devices that can test this. Apparently the issue was missed in the last round of testing. It’s a very obvious issue so it should be easy to test.
I will be doing some testing today, but messages like the ones below in the IDE aren’t giving me any comfort right now, and has just become typical behavior for ST. Ironically, these are some of the last zwave devices I have to be replaced with zigbee switches. I’t’s obvious that there is still something not right with ST’s zwave implementation, and the faster I can get rid of zwave, the better. It also gives me the option of “what if my echo +” can completely take over for SmartThings…
Sorry about the issues. If it were solely an ST Zwave implementation issue then I imagine more people would be seeing this problem. I therefore am guessing that there must be a combination of factors, one of which must be present in your setup and not commonly present in others. Have you tried to identify anything that may not be common to all users, and see if by changing that variable things get better?
Do you have a Zwave stick? If so, you can download Sigma’s PC Controller software to see the routing table of devices. I believe it shows it but I have not had much time to familiarize with the tool. It might reveal a common link between all those devices… maybe you have a bad apple which causes all the issues to devices it should be repeating for. Having eliminated most of your zwave devices may have also resulted in less routes back to the hub, or more devices going through the bad apple. I am just brainstorming… I am curious to know what has been causing these issues.
I believe you mentioned in a previous post that the issues popped up with a firmware change, but I don’t believe that to be enough to say with certainty that it is the fault of the firmware. An example of why could be that they firmed up some functionality to better follow the official zwave specifications so they can get certified for S2, etc. and for whatever reason that may have made the system less tolerant to devices that might be operating out of spec.
Everything I have left in my zwave mesh is either supported by ST, or other users have something similar. I’ve also eliminated all but a few custom DTH’s, and my zwave device count is below 30, and my zwave repairs have gone from 2 hours to 15 minutes (most with zero errors too).
I personally don’t have that, but I had access to one through an HA provider my builder uses. A while back we identified 2 older GE zwave switches with radio problems. We removed those from my mesh and put those in his environment. Same results happened with him, so those went into the trash.
The devices in a “Not responding” situation don’t bother me quite frankly because none of them are having an issue with either mobile app or automations. The offline devices do bother me. They are next to each other and within 10 feet of the hub. 2 of the switches are only separated by a bathroom wall, while the kitchen switch is only a couple feet further down the same hall as that bathroom. The last time we looked at my mesh, those switches were showing not repeating through any device other than the hub. Resolving the offline status is easy though. Just pull the airgap, toggle the switch a few times, and within a minute or two ST shows them as online.
That could be the case because none of my zwave plus devices seem to do this. I’ve also noticed significant zwave mesh stability and performance (including repairs) since my zwave device count has dropped significantly over the last 3 months. But what I can say for sure is that with one of the 24.x releases, something definitely changed back then because my zwave mesh was a strong performer until that beta I was in made another release just before public release, and then I started having issues.
Is it getting better, sure; but I just seem to have better luck with zigbee (as in 0 issues). Now that I jinked myself, let’s hope that doesn’t change
Do you have device health turned on? Mine has been off since introduction due to the issues (stuff reported as offline when it really wasn’t). I wonder if this may be a similar issue.
@schulz - I have run several network repairs with this new firmware and the log supposed to only show repair messages shows a small subset of errors or none at all, while the main even list shows them all. This appears to be an IDE issue but I did not notice it prior to this firmware release. Starting the past few beta firmware releases I can’t seem to get rid of errors… it used to take several attempts, but it appears to have gotten harder to get to zero issues reported.
Depends on the errors… Sometimes power cycling devices will fix issues, sometimes it requires removing and adding back, however the errors often go away after running the repair several times without touching the devices. I believe this may be due to network traffic issues or mere timing. Some devices are sleepy and fail to update as they do not wake up at the right time, or others may just have a dead battery.
