Thanks, but I’ve got it my possession now.
I won’t share all my USPS woes, but my post office is notorious for non-delivery and stuff showing up days after informed delivery indicates.
HA thinks it is still an aeotec 700. I’ll work through that issue later today.
RV - Wondering if you had any luck. I had more luck with the Aeotec 700 this morning but still bogging down after about 15 devices added.
I’m gonna post a more detailed description of what I’ve done and some logs, see if anyone can offer any insight.
Hi All -
Sorry for the delays, have been working with Eric (CTO) via email in trying to resolve this. He suggested a few things which we tried. I have done slightly better, but still having issues …
Eric had suggested using the alternate stick in my laptop (different port) I was successful in excluding one of the “problem” devices using PC Controller. However, then trying to include on HA failed. Interestingly, the next time I rebooted HA I had a “phantom” device show up - it was the correct next node number but couldn’t be identified by HA. I tried to “reinterview” the device and HA at that point decided that “node is dead.”
Eric recommended doing an “classic mode exclusion” on the stick. Using the reset feature on PC Controller seemed to work fine. I then the “classic learn mode/exclude” as you suggested. Put PC controller into classic learn mode, used HA to go into exclude mode. First attempt, nothing seemed to happen . I tried again, this time I did see this in the HA logs:
2023-10-10T23:58:16.405Z CNTRLR handling remove node request (status = NodeFound)
2023-10-10T23:58:16.644Z CNTRLR handling remove node request (status = RemovingController)
2023-10-10T23:58:16.664Z CNTRLR handling remove node request (status = Done)
I put things down for a few days to rest my brain (lol) and due to some work travel. I came back to it today. Used the Aeotec 700 stick on the HA setup. Used the Atrim stick on my laptop with PC controller and was able to exclude ALL of the devices (including the problem ones). That was an improvement.
Started including devices. HA using the Aeotec 700 stick. I decided to do ONLY Inovelli switches (not add in the Zooz Scene Controllers and Relay I have in the same physical area). I got about 15 devices added successfully in and around my kitchen area (mostly line-of-sight). Frankly, this was huge progress as I usually had only been able to add about 5 devices before things started failing.
As I moved further away from the control I had less success. I have a “bank” of about 7 switches near the front door. One appears to just be dead (won’t even reset). Others I was able to exclude using PC Controller but includes were very difficult. Finally got ONE to include on HA but with an “unknown error.” It wasn’t able to finish the interview with JS UI so it never recognized the Inovelli.
Tried to run a re-interview, failed again.
Tried to run a heal for the network (as some of the routes seemed wacky - i.e. going the wrong direction first). I’m seeing most of the nodes fail at “refreshing neighbor list” in the debug logs. Example:
2023-10-16T15:55:25.839Z CNTRLR « [Node 021] refreshing neighbor list failed…
2023-10-16T15:55:25.840Z CNTRLR » [Node 021] refreshing neighbor list (attempt 2)…
2023-10-16T15:55:25.847Z CNTRLR » [Node 001] requesting node neighbors…
2023-10-16 11:55:54.871 INFO APP: GET /health/zwave 301 0.712 ms - 191
2023-10-16T15:56:00.529Z CNTRLR « [Node 021] refreshing neighbor list failed…
2023-10-16T15:56:00.529Z CNTRLR » [Node 021] refreshing neighbor list (attempt 3)…
After the heal is done, I see 7 of 17 with a failed notification under “rebuild routes.” Still does not want to add devices that are physically a bit further away.
Tried to include again. Get this message:
**Node 22 added with security: None **
Expected message was not received within the corresponding timeout
Looking in the logs, it’s failing at the interview again.
So perhaps one step forward at least … but still not getting there.
Any suggestions?
RMF, not really. I’ve got an Achilles’ surgery in the morning, so I’ve been focusing on that this weekend. I’m hoping that I once I get off the pain killers, I’ll have plenty of free time to spend on HA and getting all the devices repaired while I’m on PTO.
I have one more variable, in that I also have a ring alarm. All its devices keep dropping and reconnect every several days. This is the other reason; I haven’t been pushing to resolve the HA/Inovelli issue. Leaving HA without a controller takes an excuse from Ring blaming the HA network for interfering.
Both networks have worked well in parallel for several years. If the 500 series controllers were deprecated, I’m almost tempted to get another one. but then there is the IT Manager in me that tends to push the technology envelope.
I just updated zwave-js-hi and reset one of my red dimmers and it SmartStart’ed right back. No issues adding it or interviewing it. Doesn’t help you, but it’s still working.
I do see its called rebuild routes now instead of heal.
Hey -
So as luck would have it… I got mine working today as well. No idea why… But everything paired like it’s supposed to. It was like flipping a switch.
I got an update to HAOS, HA Core and JS UI today as well… Wonder if one of those had a bug?
Do you think the update did it for you?
What versions did you upgrade from and to?
Honestly didn’t pay that close of attention.
Would this be in the logs?
I never tried it before the update, so I have no clue if it worked or not. I don’t know if I paired something on that version. There has been some broken versions released so it is possible it was the version you were running.
@PJF Yeah - certainly possible. Just was able to look - currently running Home Assistant 2023.10.3, Supervisor 2023.10.0, Operating System 11.0. I believe HAOS and Core updated yesterday. Also JS UI updated yesterday - now on 2.1.2.
By way of closure on this thread - I WAS finally able to get everything to pair up. Other than the updates listed above, I have no idea what I did “differently” that made it work on this day when it had not worked for the past 2 weeks. That said, I’m happy to move forward (FINALLY).
THANK YOU to @Eric_Inovelli who went back and forth with me over the last 2 weeks via email with a problem that doesn’t even appear to have been related to his company’s switches. I’ve always been impressed by Inovelli and things like this are why I’ll continue to be a customer and recommend them to everyone.
THANK YOU also to everyone who commented in the thread and offered me their two cents. This was a frustrating few weeks, but all of those ideas thrown together got me to the point where it’s working!
For posterity and maybe to help others in the future (as a small repayment for those in this thread that helped me) - I’ll post them below:
*My setup is a Lenovo mini-PC running bare-metal HAOS. I tried using Proxmix but wasn’t able to get it working. I personally doubt Proxmox was the issue, but now that I have it working I’m not messing with it … at least for a while lol. 3 foot USB2 extension cord connected to a USB3 port (because the Lenovo doesn’t have a USB2 port)
*Using the Aeotec 700 stick in the HA. I used the Atrim stick in a laptop running PC Controller and this was VERY helpful for updating firmwares, pairing/excluding, etc. Would recommend a cheap extra Z-wave stick in your toolbox.
*Never got the Zooz ZST39 800 stick to work properly. No idea if that was the issue, but I’ll be trying to return that sucker.
*Did a full new bare metal install of HAOS on the mini PC.
*At this point, Atrim stick went into the PC and Aeotec stick went on HA.
*On the PC Controller, excluded each device first. Then included it to the PC controller.
*Updated firmware if applicable. Both targets 1 and 2 on the dimmers need to be updated.
*Excluded from PC controller.
*Reset the switch via depressing the config button for 15-20 seconds.
*AIrgapped the switch for 10-15 seconds (note: I am not sure both of these steps are required. However, I had a few switches that after a firmware update AND reset via the config button were still not responding. Doing both of these seemed to be consistant/reliable).
-Run a final exclude, this time from HA
-Include from HA with no security
And it worked. And appears stable. And I’m done messing with it for a while lol.
Thanks again to everyone for their help and suggestions.
Just wanted to chime in and give credit where it’s due – it was @EricM_Inovelli who helped. That’s way above my pay grade haha!
@Eric_Inovelli Thanks! Too many Erics!
@EricM_Inovelli was so helpful, I owe him a beverage or three.