Can you elaborate a bit more about your setup;
If I get a bit more information we can start troubleshooting. The more information the better.
Pro 2023, Ikea Dirigera as a Matter bridge. I have about 60 Ikea devices Zigbee and matter. All my extensive troubleshooting steps are in ticket #146800 which support could not solve
So if I understand correctly you have all devices connected to the Dirigera hub and shared them with Homey via Matter Multi-Admin?
If so, is your Homey Pro connected via WiFi or a wired Ethernet connection?
Correct and I only have WiFi for homey but it’s on the same subnet als the Dirigera Hub. Pings confirm no connection issues
I wouldn’t call a simple ping to be an argument for a good connection to be honest, that simply tells us the Homey can de found on the network from the device you are sending the ping from.
How is your network set up?
Your support determined there was no network issue on my end by having me do tests.
I have a CAT6 network connected by HP ProCurve switches and a mesh WiFi 5 network.
I just got some additionial details about the ticket, and I don’t think we can fully rule out the Wi-Fi network yet.
As I understand it, you’re using a Deco M5 mesh system, with Homey Pro connected to one of its satellite nodes. The Dirigera hub is connected to the network via Ethernet.
That last part is worth double-checking: is the Dirigera hub connected directly to the main Deco unit, or is it plugged into one of the satellite Deco units? If it is connected to a satellite unit, then it is only wired to that satellite. Unless that satellite itself uses Ethernet backhaul to the main Deco, the Dirigera hub’s traffic still depends on the wireless mesh connection between the satellite and the main unit. In practice, that means it is not truly wired all the way back to the main network.
This setup can mean that Homey has to communicate over multiple network links before its traffic reaches the rest of the setup. The main Deco needs to maintain a solid connection to the satellite, and Homey also needs to maintain a solid connection to the satellite. If Dirigera is connected to a satellite without Ethernet backhaul, its connection also depends on that same wireless mesh path.
For general use, such as browsing or watching videos, this kind of setup can appear to work fine. Those use cases can usually tolerate brief buffering, retries, or small delays. Matter control is less forgiving, especially with a large setup of 60 or more devices and Matter Multi-Admin involved. Small drops, packet loss, or delays can have a much bigger impact there.
In the diagnostics report, we can see that the connection is being lost. After that, Homey tries to reinitialize the connection, but that process takes too long and eventually times out. That suggests the issue may not be the initial pairing itself, but the stability and recovery time of the network path. If the connection drops even briefly, Homey needs to re-establish communication before it can continue. If the network does not respond quickly and consistently enough, that recovery can fail.
You mentioned in the ticket that you tried connecting Homey to the main Deco unit but did not notice a difference. How far away is that main access point from where Homey is placed? It could simply be too far away for Homey to maintain a stable enough connection, even if it technically connects.
My suggestion would be to verify the actual Deco topology in the Deco app: which node Homey is connected to, which node Dirigera is connected to, and whether the satellite Deco units are using Wi-Fi backhaul or Ethernet backhaul. If possible, it would also be useful to temporarily test with Dirigera connected directly to the main Deco unit and Homey placed close to the main Deco, just to remove the mesh backhaul from the equation.
Thank you for taking the time to look into it.
Homey is connected to the main (loced to that unit only) unit as is the Dirigera unit (by ethernet so it’s not using the backhaul.
Homey is now 4 m away from the main unit, but also tried them closer to each other.
Im posting this to hopefully help someone who finds this forum or is new to this…it may not be so useful to others of you who are more advanced and I cant guarantee this will be a fix for you.
I know this topic is for Homey Pro (im shs) but it might still be useful because like me people will find this topic when looking for help.
There have been many posts and discussions about Matter /Thread / Ikea devices on matter and I cant remember where i posted in all of them so some of what i say may be repeated or irrelevant on this exact discussion. However there may well be some who are still struggling or who are new to Homey that find this.
I am on SHS. I had Homey Pro 23 but memory was an issue so i moved to SHS on its release. Matter was never strong in the HP23 (much of which is probably attributed to what ive learnt and said below), i had many drop outs. When I made the jump to SHS it all fall apart
I could not for the life if me get my Thread Nuki lock or Thread devices to join Homey.
Ive been a SmarThings user since their first days, tons of my devices lived in ST and so that was as my main TBR. This was extended using some Onvis plugs and an Ikea hub which I changed to share the SmartThings Thread credentials. Everything went into ST easily (except Nuki) but i could not then share them via matter into Homey. If they did join, they would go offline after a few hours. This was painful
Eventually, having submitted a few tickets I was told they could not help and to ask the forum. I asked a few questions and found many had similar issues. So i gave up and went to Home Assistant which I used to bridge my devices over from SmartThings. This worked perfectly so I sort of gave up and left it there. As we have seen there has been many conversations around this and myself included vented some frustration at the lack of support ( lets not get into that now!)
In the last few weeks with the news ST will be adding a paywall, i wanted to shift my devices over to homey. I still had 30+ zigbee devices in ST due to the limitations of the Homey Bridge and I wanted my Thread devices in.
I sent a few messages to @Doekse to see if he could offer any insight because I obviously wanted my Nuki and TRVs in Homey.
What worked for me and what ive learnt
I learnt that several thread networks can cause an issue. I didn’t realise that Tado Wireless receiver (which drives the boiler when a trv asks for heat) pushes its own thread network. this is locked down - nothing in the thread network can be changed at all and it cant join another thread network. So i had this and my smartthings thread competing with each other - I “thought” that my ST hub was the main TBR but it wasn’t.
Homeassistant was able to show me the preferred network - this was Tado.
Firstly I wiped and reset my ST hub and factory reset all my tado devices.
Credentials
I had previously noticed that using an older iPad helped Me get devices into Homey. Another user on here or another discussion has said the same. Again, i was unaware that the credentials of the preferred thread network are stored on your phone or device. This would make sense as to why I couldn’t get devices to join with one mobile but perhaps can with another and also why they were unstable.
I deleted all the devices from my matter menu in ios.
I also purchased a SLZB MR5U to handle the Zigbee - this has 2 chips, one which can handle also thread.
I then made ST, IKEA and the MR5u all join the Tado Thread network. (All on ethernet in different rooms)
This made one unified thread network. I used home Assistant to send the credentials to my phone.
I was then able to add every device into homey, no issues at all. They have (touch wood) stayed connected and responsive. I started with the powered stuff to help the mesh and then left It a few hours, i then added the lock and TRVs
In a nutshell, its worth going through your network and your devices and it might be worth the time starting over. Yes this has taken me a few days but
Nuki Lock (pro 4). I had many many issues with Nuki in all platforms. their support were awful. one user on the SmartThings forum advised me to turn off the built in wifi in the device. That worked, the lock was stable in smartthings and now homey without built in wifi.
This might nit help you all. This might be a waste of time but hopefully it helps someone or makes you re-think your set up like I had to. Even if that means a few days of head scratching and device resets
I have made suggestion that Homey has a better way of showing Thread networks (not possible with shs at all) because the home assistant thread tool is what really made me see the issue.
Edit - update the Thread button in the app for SHS users has been removed but will be back in next app update apparently. Hopefully this functionality improves for all Homey users over time
Hello @Doekse,
I know basically i should report my problem to the support team what i always do in the past, hopefully its okey to go this way if not please tell me and i open my last Thread Problem Ticket (150489) on the hotline again…no problem!
Diagnostic report number during the Thread-Matter crash and before Homey Restart: 8CD746BC
The report was generated before the Homey restarted. The problems started today 17.08.26 at 7:14 PM seemingly out of nowhere. Even after waiting four hours, a stable Thread-Matter network could not be re-established. Out of 26 devices, a maximum of 11 and a minimum of 4 were offline. The network topology fluctuated constantly between these failed devices.
It began with the failure of a single device, Node =0xe84cd98f=Meross Presence Detector, (Image 1), followed by further failures (Image 2), which shows the timeframe of the last seven hours.
Here you can see repeated attempts to rebuild the network, but without success. The Developer Tools show that different devices fail each time and lack an IPv6 address. Image 3 shows a recap of the last 7 days, where you can see there were no problems.
until today…
My setup:
I’d appreciate your input. I already contacted support about this issue and sent several reports and screenshots. An ATHOM technician even had direct access to my Homey, but unfortunately, nothing much came of it.
I then decided to set everything up from scratch directly through Homey, deleted all multi-admins, and bypassed Apple. Apart from a few minor issues, everything ran smoothly until today. Out of nowhere, suddenly, nothing worked anymore, without anyone having changed anything.
Should I restart the Homey, or would you prefer an ATHOM technician to take a look at the problem directly? Should I open a ticket? Please give me some brief feedback on the best way to proceed. In the past, restarting Homey resolved the issues, but now we have the opportunity to get to the root cause, right? Thanks for helping!
[/grid]
well…today the new firmware was released 13.4.1 and there are changes regarding matter, maybe it helps…12 hours later no improvments in my thread network, on the contrary it got worse see screenshot attached…i will install now the update und reboot the system, this is my diagnostic report from today- 8ED30B21 which i made before i installed the update, i will forward it again to the support team, hopefully they can figure out what caused the thread-matter crash…
you can try to get additional logs via this app that I create [APP][Pro] Thread Diagnostics - Continuously log your Thread mesh & Matter device health - #5 by staeff
AFAIK the antenna design is the same as the 2023 model. The dual protocol firmware is not ideal, but for running 30 devices (without actually using zigbee) is pretty much the same setup I have and it has been running stable for months for me.
thx for the feedbacks…i know the topic with the antenna/chip, this was already on the table in this thread…also my system run stable over one month…what i want to figure out, how is it possible that e system run stable for weeks and at once, like you flip a switch, nobody change, anything and the system implodes and fails to come back…only a reboot solve the problem at all till to the next crash - days-week-month…?
well, i have again report now all details to Athom, Support Ticket now - 153766, they receive now 3 reports from all situations
1 - 8CD746BC during Thread-Matter crash active
3 - 8ED30B21 12h later Thread -Matter crash active
4 - D59B7E31 after reboot the system - Thread - Matter stable back again
A lot of screenshots which are named by the situation and a detail history description what i did, whats happen and which devices are involved.
Hopefully they can figure out something, if i get the feedback i let you know.
Mine had been more or less stable for weeks. But this morning I opened a window and the roller shutter did not move up.
I checked the Homey app and the window contact was still showing closed. About three minutes later, the roller shutter suddenly moved up and the contact finally changed to open.
Seriously, three minutes?
For something like lighting this is already annoying. For anything even remotely safety-related, it is simply unacceptable. I am planning a Flow that opens all roller shutters when a fire alarm is triggered, so that possible escape routes are accessible. A delay of three minutes in such a situation could have serious consequences. Obviously, I would never want a smart-home automation to be the only safety mechanism, but if the automation exists, it still needs to react reliably and within seconds, not minutes.
Then I noticed that one of my Bosch TRVs was offline as well.
Since I disabled automatic updates, I saw that a new Homey update was available and installed it manually. Afterwards, some devices needed almost 30 minutes to become available again. With Apple Home, rebuilding the Thread network after a restart typically took only around 3–5 minutes in my experience.
And the Bosch TRV was still unavailable, so I had to go through the repair procedure.
This is what keeps frustrating me. Every time I start thinking, “Okay, perhaps it is finally stable now,” another problem appears.
With every new disappointment, I am getting closer to moving to Home Assistant in October. I already have a large number of devices and many more are planned. I simply cannot continue building a larger smart home around a system that is, by far, the nicest and most pleasant one I have ever used, but whose reliability I still cannot trust.
And whenever I submit a diagnostic report, the answer is essentially the same: nothing particularly unusual can be found.
That is becoming difficult to accept when the real-world symptoms are so obvious.
While this can be a good solution, please do not rely on Homey or any other smart home controller for mission-critical devices. A smart home system is not designed for that, it’s designed for convenience. Smoke alarms should always be able operate independently of any smart home system
I am not an expert Matter/Thread but AFAIK it is based upon TCP/IP network technology, which after so many years of development and improvements has shown that a stable network is possible and can deal with a lot of situations. So I think it is scaring that Homey cannot do the same. Even more scaring that even the other network technologies Homey is using can just fail out of the blue and that the failures are not solved by the software it self. So what is going on?
The hardware is not futureproof
Well, my Homey 2019 supports all cloud, BLE and Zigbee devices perfectly without any issues. So I don’t think it’s a hardware issue.
