SJK

I had an old thread on this topic, and am still having the problem. With the latest eISY OS release, I saw a note under fixes "Zigbee devices could not be removed". Wondering what was fixed specifically. I am still stuck with 3 Zigbee nodes that I cannot remove from the Zigbee network and yet again, the only solution will be to rebuild my entire Zigbee network (and then nodes, programs, UDM buttons, etc) for the 4th time as each time one of a node becomes unresponsive in the network they never get removed when removing before deleting and then I'm stuck. It's getting old- I'm about to resort to a Hue hub to at least protect that part of the network then figure out some other solution for the rest of the nodes - if I could figure out how to reliably integrate eISY with a separate Zigbee controller.

I didn't want to pollute the new release thread with discussions out of the scope for release troubleshooting, so I am starting a new thread. I like all the new functionality for programs- haven't started building/modifying programs in the eISY UI yet though but excited with how it looks. While reviewing, I thought of a helpful feature to speed program debugging- showing true/false status line by line in the program conditions and whether or not the then or else clause is evaluating to true. This should be doable as on the program list you can see true/false status and on the variable pages you can see their values. That would be a big visual clue as to the error in your logic or confirming it's working as intended. Otherwise you have to open multiple browser windows or flip back and forth between pages to step through a program. Secondly it would be great to be able to eventually create variable widgets on the nodes pages like I can with UD Mobile. I don't use the eISY UI at all to interact with my systems because I don't generally directly manipulate nodes - I use state variables for many things to track button presses around the house or desired settings. Example- I don't ever touch a thermostat- each room in the house has a desired state for mode and temperature that is determined by many variables, such as occupancy, home occupancy, window and door status, outside temperature, season, forecast, etc. If I directly change a thermostat on one of these interfaces without first setting an override variable/flag, eISY with just put it back. I do this type of setup for most every subsystem. For lighting, eISY tracks room occupancy based upon motion or manual button presses and runs timers. If I want to turn off all the lights in a room, I'm not turning off each node- instead I've used a UDM variable button action that just sets the occupancy to zero for that room when pressed. And another timer kicks off for HVAC setback in that zone as well. On mobile I can include these state variables and desired settings right along the thermostat display widgets on the favorites section, for example. Then every control surface could just show the nodes page rather than having to update my iPhone UDM build, then export/update to my spouse's phone and kitchen iPad and also track any changes in parallel with eISY UI too.

My upgrades went well. On one eISY I had to reinstall all my plugins, the other one was fine (they are the same plugins). Noticed on one eISY (the one I did not have to reinstall the plugins on) is showing the wrong time zone (2 hours ahead) in program last run/next run and variable last updated views. The PG Log (which I am using because I can see it in real time on the site) is showing local time in the one working properly but 2 hours BEHIND in the one with the TZ issues. However on both machines in the AC the time and time zone is correct.

Same issue here. I'm getting update available notifications on UDMobile for both of my eISYs and the same screen in the update firmware. I'm on 6.0.3 and i says upgrade to 6.0.0. In the AC, the release notes say to use UDMobile to upgrade, the dialog says to simply press upgrade packages. Which is correct or does it not matter?

Sure- they are a basic Zigbee3.0 device. They are nothing fancy but work pretty much identically to Philips Hue bulbs. You get an on/off node, an x/y/temp node, and a color/sat node. All work exactly like any other ZB3.0 device. My only con is that if they are reset (or powered off for months, like happened with my exterior flood lights) you end up with a new ID (think this is typical of these kind of devices) which makes reprogramming a pain if they are in multiple scenes. Always remove from network first of course. I recommend these for LED strip drivers. They have few models with varying configs. They are inexpensive- no reason not to get the 5 in 1 version that can do RGBWW+CW for strips. They can be configured to whatever strip type you need- RGB, + WW, +CW, etc. They even have constant voltage modes and single color dimming modes or CW/WW modes. Pretty flexible- cannot go wrong here. Just pick the device that meets your needs.

I have many of these GLEDOPTO LED controllers. They make a handful of zigbee models. They work fine out of the box.

Doing some forum searxihng about orphaned zigbee devices I found an acknowledgment in 2023 that this was a bug and would be fixed within the week, followed by a user stating still not fixed a few months later. Other threads were all closed without any replies or that upgrading firmware fixed it. Perhaps a regression? I'm starting to think that if the controller still thinks the device is there, it's not going to re-pair. But if I cannot delete it, I'm stuck. I'll put in a ticket to ask.

I did factory reset them multiple times - screw/unscrew method. They were successfully reset. Turn off the bulb for 5 seconds Turn on for 8 seconds Turn off for 2 seconds Turn on for 8 seconds I have another one of these bulbs planned for another room when I install the pendant- I just tried to link it to the eISY in place of one of the nonresponding bulbs and it instantly paired, confirming it's not a radio interference issue, etc. When I use the Xray function to see the devices in the ZB network- I have 16. I only have 10 devices on this network. There's devices there I don't have and the ones that don't respond and have been removed from the eISY. None function- I get errors if I try to interact with or query them. I have removed them from the zigbee network before removing them from eISY, but they still persist in the all devices table, and if you do a synchronize new and delted with interview they come back. How do I permanently remove them from the network? Not sure this has any actual bearing on the actual problem, but seems odd I cannot remove anything ever.

I use some Tuyo clone things for window candles- the ones that you splice into the power cord. Something like those old Sonoff Zigbee basics. Also a couple of Sonoff USB on/off switches for the Lectrofans. All of those items are working just fine as well. I use no hub or nodeserver for the hue devices- they are linked directly to the eISY using the ZMatter dongle. Same for the other devices mentioned above.

I have a handful of Hue bulbs (and some other Zigbee devices) on my eISY. Within the past week 2 of the hub bulbs stopped working. These are the fancy lightguide glass bulbs. First the one in the master bedroom pendant stopped responding, then a day or so later, the one in my daughter's bedroom pendant. I factory reset them but they will not re-add to the eISY. First I tried an update/interview, then the reset with removal from the eISY, but again, will not re-pair. The lights work with bluetooth on my phone. (They aren't dead). I set up the hue app to verify that they still function. They are only maybe 2 months old? I doubt the zigbee radio is dead- they failed nearly simultaneously. No power failures or surges etc- I use whole house surge suppression and we have relatively "clean" (stable) power here. The only interim change is upgrading to 3.9.1 and turning on Matter support. I have no matter devices, and I did try turning it back off without any luck. But this issue didn't temporally appear- I updated to 3.9.1 a week or two after it came out, and this issue started a week ago. My zigbee network is fine- other devices work, including the other hue bulbs in the house. The regular 60 and 75w A type bulbs are still doing their thing- including those in the same rooms. This isn't a range or interference issue- nothing else has changed and again, everything else just works. (Side note- have the Polisy running a separate Zigbee network on a separate channel for exterior/window devices- completely separate radio frequency, should not have any issues there- have been happily co-existing since the fall). Any thoughts? I have no idea what to troubleshoot next.

not enough info in the post to troubleshoot, but here's a few starting points: 1) You have the towel bar 2 hour off program currently running while towel bar AM is also true- check out why (if it's a simple logic of if towel bar on, wait 2 hours then off, that would explain it.) 2) Check the date/time/timezone at the top of the AC to ensure it is still set correctly 3) Are the devices themselves responding? Turn it on/off at the device and see if it works manually and that it's status is reporting correctly in the AC Reply back with that stuff first (Is it plugged in, etc)

The new 5.9.1 release and the instruction/admonition to ensure you have good backups before proceeding got me thinking. Do I understand the complete backup strategy? I didn't find the complete answer here so thought I'd ask. Obviously there's the "backup IoX" menu item. Used to take a long time- now it's incredibly fast and makes a small zip file- too small... Makes me wonder what's not included. What does the backup IoX file include? All programs? All configuration tab items? Per the migration guide it seems like an all-in-one backup. But IIRC my migrations in the past were not smooth (but there were version and hardware differences- was not a straight like for like restore) requiring a fair amount of re-work. Topology file? (Good to have, not a restorable item but contains some very valuable information) PLM links table? I've found a few of those files lying around my backup folder and cannot figure out how to do that again if needed. As above- do I backup my programs separately? (Even if included in the main file)- a good idea in case a partial restore is needed due to a programming misadventure? If I export the root folder of my programs- are all included or just that root folder? File size seems really small given the amount of programs I've built. Variable definitions? Main file I'd expect? Z-wave- don't use it but see there's a separate procedure to backup a Z-wave network. What about ZigBee? Matter? Polyglot- has it's own backup system. Any caveats there? UDMobile is relatively straightforward, obviously a separate step. I have multiple production devices- no spare equivalent test devices- not interested in blowing up my controllers to f-around and find out on my own if my backup is complete at this stage. What does everyone do?

More sonoff devices- Water leak sensor is recognized as a water leak sensor class device by IoX with multiple parameters but water leak detection events are not received. https://sonoff.tech/product/gateway-and-sensors/snzb-05p/

Sonoff ZBMINIR2 on/off switch. Currently able to be added as a Zigbee device, on/off works, but no ability to configure/mange device settings such as detached mode, inching mode, trigger mode, power reset state. My use case is to sense a local switch for Hue bulbs to toggle on/off state without powering off bulb entirely (originally wired it as a local failsafe switch/if we ever sell the house and revert to "dumb" wiring). A whole lot cheaper than another insteon switch with a disconnected load. https://www.zigbee2mqtt.io/devices/ZBMINIR2.html https://sonoff.tech/product/diy-smart-switches/zbminir2/

I currently have zero Z-wave devices at this location so could live with that. Too many scenes with each devices a member of multiple scenes to accommodate on insteon PLM. Attached article not too helpful- one network manager per network- but I'm looking at making multiple independent networks - or at least that would be fine as I see no need to make both see each other. It does mention "start network and join/leave network" so presumably each controller should set up their own independent network - otherwise how would interference between neighbors be prevented? Found this: https://www.rfwireless-world.com/Tutorials/Zigbee_tutorial.html which mentions each controller starts its own personal area network- so I can see on my eISY this in X-ray of the network: Integer DATA controller.data.defaultPanId = 27238 (0x00006a66) I would assume that if I bought another ZMatter dongle it would create a different network ID- either randomly or upon detecting a collision with an existing network but that is a something one of the UD developers would have to confirm.

As an aside- just received my first actual Philips Hue bulbs and tried one out. Amazing that I can set commands directly with scenes included alongside Insteon switches- so I can use the regular wallswitch button to turn the light on AND reset the temp back to a "normal" bulb rather than the last on. Well done native implementation! Kudos!

I don't have a large zigbee network, I have a large Insteon network. The Zigbee devices will be in locations covered by each of the IoX controllers (max nodes/scenes/devices per PLM) and cannot be reconfigured to condense the Zigbee network to one IoX controller.

I have a corner-case question here regarding Zigbee (and possibly ZWave). I have an eisy and a polisy both running in the same home as my Insteon network is too large for one to contain all nodes. The polisy handles exterior and basement, the eisy handles floors 1-3 in the main home and the entire HVAC system. Polisy handles polyglot and external intgrations and the eisy references polisy for polyglot. I'm now adding Zigbee to my home- currently for Hue-style lighting using the GLEDOPTO 5 in 1 LED drivers and their exterior wall wash and garden lighting (integrated Zigbee 3.0). Problem is I'll have some inside the home, where it will be important for them to integrate with the interior scenes and programs on eisy, and also the garden/wall wash exterior ones which should be integrated with the exterior scenes- on polisy. Since z-matter works without a hub I don't have one so I don't need or want to use polyglot/nodeservers. That would make this easier as both could "see" the nodes (like I do with Elk). I have no problem getting a second ZMatter dongle for the Polisy. But then I have two controllers. Can they be on the same network? Should I just change the frequency of one of them (no idea how to change the frequency of the listening devices or if they even support that)? I don't have any way I can reconfigure which IoX controls what given the HVAC is all on eisy and I've previously run out of Insteon nodes/scenes and my final insteon network plan has gotten substantially bigger since then. Thoughts?

On the device dashboard for a variable the top 3 pickers are under the properties header- value precision and startup value. Value is where is get that error- tried with variable dashboards of status commmand and default types. Below that under a command header is a blue button for value which will accept an arbitrary value. Since the property picker is first (and there and accepts a value) it’s what I always click on.

Sorry if this is double post, but I can't find it with search so I'll ask the question here. I have a few minor issues with the mobile app on iOS: When I go to manually input a variable value in a entity's dashboard, I cannot enter a value greater than 100. Obviously in the admin control panel I can enter any value I wish. Obviously if you were looking at status of a light entity a hard limit of 0-100% would make sense but I'm talking about a integer or state variable which should be able to hold any value. The value can clearly be something other than 0-100 because I can view variables set via my programs with arbitrary values. Similarly I cannot set the value from the variables list directly outside of an entity/device dashboard. If I create an entity/device of type command with dashboard, the ability to edit the commands at the bottom of the pane disappears. I can add new steps but cannot edit existing steps. The little right arrow/Chevron that would open up the command to edit literally isn't there. If I change the entity to type command then the Chevron appears, and I can edit a previously entered command step. A question on trying to create a button that when pressed would increment a value. Due to how I've laid out the pane in question, I don't want to use the increment button type. I'm not trying to manually adjust the value directly to an arbitrary value- It's really that I want to press the button and have it add 30 or 60 (minutes) to the existing state value being used to feed a timer program. The command entity cannot add to an existing value (+=), only assign the value directly (=). Am I approaching this wrong or is this a limitation of the app? I could write an increment program in the admin console and have that button run the then step of the program I guess but that seems like a workaround. Overall, kudos with the app. It's really grown and has become indispensable to running my home (as much as I've automated everything to require minimal interaction.)

No. It’s an API call to a REST URL with parameters. GET/POST command with the inbuilt network module will suffice. I use IOX variables to hold state variables for status and integrity variables for the desired setting prior to the call.

I have noticed that occasionally some nodeservers stop working on my eISY's IoX admin console. What I've traced it to is the nodserver profile address on the IoX service. Polyglot is working fine- the nodeservers are fine, but IoX changes the URL/IP of the nodserver to 127.0.0.1. I run both polisy and iox due to running out of Insteon nodes a while back so I split the install. The polisy is older and is already running polyglot so I kept it there. I do not use the eISY's polyglot install. Instead i've added both ISY mac addresses to the polyglot server on the Polisy and run the required nodeservers there in one or both profiles as required. That cuts down on licensing and keeps everything in one place, simplifies updating, etc. It works fine and seems analogous to when polyglot first came out and I'd run it on a RPi with an ISY/99. But if there's a network hiccup, or I update a nodeserver (so it goes offline) the eISY will often revert to a local address, breaking the connection. Oddly the nodeserver is still listed as online in its node but control breaks. (Like the Elk module will pick up events - violations, arming, etc. but I cannot send any commands from IoX). Go to the profile and put in the correct IP address and everything is fine again. This is longstanding behavior- as long as I've split out to two systems. Note I haven't figured out how to completely turn off the eISY's polyglot service. I don't see an option to do so. I could write a script to disable the service but I'm sure a future update will revert that or break a version update or something so I'd rather not go hacking around in a command line. But the eISY polyglot isn't running anything- no nodeservers, etc. Previously I'd be able to log into the eISY polyglot -not that I need it for anything thought- but I just tried while writing this post so that I could again confirm it's idle. I couldn't log in via the website/port 3000. So I logged in via the console and checked the logs: Jan 18 17:05:04 eisy daemon[2932]: 1/18/2024, 17:05:04 [pg3] ^[[32minfo^[[39m: call_udx: Making request to socket with url /rest/pg3.auth.ns Jan 18 17:05:05 eisy daemon[2932]: 1/18/2024, 17:05:05 [pg3] ^[[32minfo^[[39m: call_udx: Making request to socket with url /rest/pg3.auth.ns Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: Verifying node servers are installed on IoX correctly Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: IoX Request: [Try: 1] [00:21:b9:02:64:d7] :: - http://127.0.0.1:8080/rest/profiles/ns/0/connection Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: IoX Response: [Try: 1] [00:21:b9:02:64:d7] :: [200] :: 3.1233ms - http://127.0.0.1:8080/rest/profiles/ns/0/connection Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_1] Unmanaged NodeServer 'UnifiPresence' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_3] Unmanaged NodeServer 'WirelessTag' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_2] Unmanaged NodeServer 'ELK' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_4] Unmanaged NodeServer 'Timedata' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_5] Unmanaged NodeServer 'Holidays' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_6] Unmanaged NodeServer 'Sun' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_7] Unmanaged NodeServer 'OpenWeatherMap' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_8] Unmanaged NodeServer 'Virtual' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_9] Unmanaged NodeServer 'Roku' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_10] Unmanaged NodeServer 'Plex-WebHook' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_11] Unmanaged NodeServer 'VenstarCT' found. Adding to DB. Jan 18 17:05:06 eisy daemon[2932]: 1/18/2024, 17:05:06 [pg3] ^[[32minfo^[[39m: [00:21:b9:02:64:d7_12] Unmanaged NodeServer 'NOAA' found. Adding to DB. Jan 18 17:05:08 eisy daemon[2932]: 1/18/2024, 17:05:08 [pg3] ^[[31merror^[[39m: UDX authentication failure for admin with cause udxTimeout: Error: udx /rest/pg3.auth.ns - ECONNABORTED: UDX timeout after 4000ms Jan 18 17:05:09 eisy daemon[2932]: 1/18/2024, 17:05:09 [pg3] ^[[31merror^[[39m: UDX authentication failure for admin with cause udxTimeout: Error: udx /rest/pg3.auth.ns - ECONNABORTED: UDX timeout after 4000ms Never seen that before- every 5 minutes it's polling IoX and adding nodes to the internal database. Never saw that before. Last night I upgraded packages on both polisy and eISY to get to IoX 5.8.1. Thoughts?

One more thought if this is relevant to your decision: Venstar API is local. I am free from and service/cloud dependency. I'm not opposed to using one, but we all know services shut down or change their business models, nodeservers get abandoned, etc. Skyport is there if I wanted some cloud-baed vendor-supported tool, but separate from the API access which is on-device. Isn't ecobee reliant on a cloud connection to their servers? I've never used them so I could be mistaken, but when I was looking at them I recall that was an option- they have a cloud service which they are pushing people into to get some "security" functions out of the sensors (motion, window monitoring, etc). That required a subscription to enable and I didn't find that appealing. Also if you do go ecobee- look at your utility provider- many have discounts for purchasing a "smart" device through them and lots partner with ecobee or nest. Pay up front, pay later, or pay with your data.

I vaguely remember a scheduled reboot feature in the old ISY. I cannot find it on IoX now. I wanted to try a scheduled weekly reboot because I'm having stability issues lately. eISY will hum a long for weeks then all of a sudden everything goes wonky- variables and program status seems off and things stop responding properly with no changes in the setup. I have a lot of programs and variables including currently 223 state variables and 295 integer variables. I'd think eISY can handle it but perhaps there's a memory leak somewhere. In any case a reboot seems to set everything straight for a while. Is the option for a scheduled reboot still possible somewhere?

Also look into notifications via UDMobile. Much more reliable and there's a wiki post to help you configure the network resource.