I found one more curious correlation with the first of my two CPU temp spikes. Weirdly the amount of free storage in Homey seems to have gone up markedly at the exact same moment as the start of the first CPU temp spike. I made no change to Homey e.g. installed/uninstalled apps at that time to explain it. No such correlation with the 2nd CPU temp spike though.
Complete speculation based on very little understanding of how these things actually work but I wonder if Homey cleared out some of the very large Swap file which I had before I did the CM4 mod?
On my side I haven´t found a pattern yet - sometimes I have 3 unexpected reboots on a day, and then, my homey runs for 3 days without a reboot, but always with those flow-effects.
Well, somehow the Homey Web interface when updating Advanced Flows got very slow, not updating so fluently like in the past and yes, I observe it already longer time.
I have the same issues as described here on my homey 2023. Got desperate and bought the 2026 version to get the exact same instability. Currently having such remorse that I gave up and are migrating to Home Assistant. This however is quite challenging as i have many complex automations. So today I start migrating first to homey SHS to see if it is hardware or software related.
I did make a support call to homey, but the response was so disappointing that I gave up immediately. I think it was totally unacceptable.
Together with long lasting Zigbee issues, moving to HA seems the only way out on the long run. Sad that this product is still to unstable to really rely upon.
Not sure how everyone else is getting on but stability has got worse not better for me over past few days. Previously I had a couple of random restarts a few days apart. Now I’ve had 3 in 36 hours. Only thing that changed was I let Homey do the update to 12.11 stable firmware. No correlations I can see with that or anything else. Two of the reboots have been since then but the first wasn’t for quite a few hours after the firmware upgrade.
I’m still seeing random periods when temperature rises by ~5C sustained for several hours then drops back down again but these are not correlated with CPU load changes and I’ve now been able to disprove my earlier hypothesis about them being linked to the LED red ring showing (it isn’t; LED ring is staying off throughout).
Homey’s job number 1 is to be a rock solid place for my automations to just work and it’s failing on that front now and with no clear route to it ever getting better. It’s all a bit sad.
It happened to me already as well and yet, day after it was again stable…so wait another day. After last update it’s not worse however, it’s still random.
So far I haven’t seen the correlation, which would point to the common ground though
And now this evening a whole new nightmare. It seems Homey has now lost contact with a whole load of my Zigbee devices and none of them can be repaired exactly is things were when I first migrated. This includes all my radiators. The “fix” for that (which obviously didn’t last) was to reset my Zigbee network and spend half a day repairing all the devices.
Homey is rapidly becoming completely unusable. I am forced to replace it with something otherwise I cannot control my house and I don’t have multiple tens of hours to put into trying to get it to work again with no expectation that any of the fixes will last or be stable. This really is catastrophic for my Homey experience. I didn’t think anything would push me away to another platform (if nothing else for the extreme high cost in time of doing so) but this really might.
Well, a pity you did ask before performing the reset, but just by any chance, what Homey Developer Tools was showing in the field Current command and Firmware version ? Frankly, it’s not normal and very rare seeing Zigbee mesh would get destroyed just like this…
But please note this has probably nothing to do with the random reboots/instabilities we are facing here.
Btw despite instabilities, I’m running with 82 Zigbee devices with Homey…
I didn’t reset Zigbee network today - that was the fix I had to do a few weeks ago to overcome the problem where Homey is incapable of restoring the backup of a large Zigbee network. Since these random reboots I have done nothing to Zigbee (other than note that tonight all the same devices that were broken when I restored from backup in that previous adventure were again broken tonight).
Sorry, I’m getting lost here. Anyway, instabilities (unexpected reboots) as such are not influencing Zigbee / mesh stability. And yes, if you were restoring eg. from HP23 to HP26(CM4 swap?), immediately after the restore you might have noticed that Zigbee might be not alright. But that’s not itself necessarily problem of Homey, could be specific to the devices you are using. Previously you wrote about CM4 swap for that the best is to perform USB backup and restore, not the cloud one but again - I did the CM4 swap recently as well (to exclude HW issues of previous CM4) and all my 82 Zigbees are working fine, I didn’t have to touch anything since then….
Yet, I’m also user of Zigbee2MQTT in parallel to my Homey - it’s also not absolutely without problems… but that’s another discussion.
I think that is because you did not change to a different Zigbee chip. It’s exactly the same. You only changed the CM. When migrating to another Homey device (so with a physically different Zigbee chip) the issues mentioned in this topic may arise, when you have a larger Zigbee network.
Yes that’s right, I was there in the topic you linked to, discussing it. That was a few weeks ago. After I did a Zigbee network reset and repaired everything, all seemed fine for several weeks. Until now.
I’m not sure why you’re confident the re-emergence of this issue has nothing to do with the reboots. Nothing else on my system changed in the meantime and the re-emergence of this Zigbee issue occurred after one of my random reboots. I don’t know whether it’s related and I don’t think you do either. I’m just trying to contribute all the behaviour I’m seeing associated with these random reboots in the hope that some help can come.
I’m frustrated because I rely on Homey for my house to be warm and lit so when it randomly decides to break in multiple ways repeatedly for no reason, that’s really quite annoying. This being a user forum, I’m not complaining at other users, more commiserating with my fellow sufferers.
Can you actually confirm the data I showed it the first post of this thread :
That would be some confirmation you have similar cause for reboots like me, unfortunately haven’t seen it.
Second, re being “confident”, I’m not - but to troubleshoot ZigBee issues, there are many many steps you shall normally do to confirm the last options is related to the random reboots - it would be yet would be very odd (during the crash the homey-pro is terminating, the zigbee demon finishes in controlled manner so I would not see any relevancy, but…anything if possible).
eg .
- reconnect all non-battery powered Zigbee routers
- check if you don’t have Wi-Fi interference, eg. could be neighbors could be your router autoselecting channel etc
- ensure in case you have ZigBee lights, they are not powered off
- ensure you have enough ZigBee routers
- avoid some devices like particular Ikea ones
- unplug Homey fully from electricity
- also check the power supply or USB cable (I myself experienced faulty USB cable just based on inability to perform USB backup)
etc.
Yet, you start directly repairing ZigBee…have you done/check any of above?
.