Offline again. Log number 1649280 @WyzeJasonJ
I’ve said all this before. I have two thermostats running the same version of firmware on the same tplink mesh network at my house. Only one of them drops offline. No other Wyze device I own exhibits anything like this issue.
I just wish the Wyze would devote the resources to find and fix this issue.
Not sure if you have tried this. Judging by your comment that the firmware is the same, I can assume both thermostats are the same. Have you tried swapping them?
If you have what we’re the results?
If not turn off the breaker or switch powering the furnace.
If your tstat is same as mine you should be able to pop the unit off its base plate.
Swap them.
Power up the breaker/switch.
See if the problem follows the tstat. If it does you probably have a bad unit.
If the problem stays with the location then I would suggest you have a power or wifi problem.
One of the troubleshooting problems with doorbells and tstats is you can’t move them an inch or two and see if things change (yes, you can with a lot of trouble).
Final thought. I’m a bit wary of wyze’s 3 bar signal quality indicator. If you have 3 bars I’ve found that to be a good thing. If you see 2 bars it may or not be good. I strive to get all 40 of my devices at either 3 bars or other vendors 4 bars.
Keep us informed.
I’ll gladly make that happen and report results.
Maybe you’ve been following this thread for a long time and have heard the dozens of folks here talk about this ongoing issue. Maybe you just thought of this as a possible cause and decided to share. Or maybe you are new to the list and haven’t heard all the things that have been tried over the last 3-4 years after a firmware update appeared to cause this to happen.
A side note: one firmware update around that time caused a thermostat to brick if you reset the device with that firmware installed. I know because I had one of those. I submitted a request to WYZE support saying “I am hearing that this firmware causes the device to brick when you perform a reset”. WYZE replaced the unit without question, but also without saying “this is a known issue”.
It’s certainly possible that all of us have bad units. HOWEVER, that would mean there are a lot of bad units out there - certainly possible. It would mean that the fact that this problem in a thermostat at my mountain cabin which cleared when I disabled WMM in the router at that site (over two years ago) was a coincidence. If this were true, one would hope that WYZE would just say you all have bad units, rather than having UA continue to submits logs into a black hole.
We’ve all been sending logs to WYZE for a l-o-n-g time and have been waiting for some response from them. “This is caused by bad units” would be a perfectly acceptable answer and we’d all gladly surrender these defective units for new ones.
Sorry to go on about this, but we’ve all grown very tired of waiting for an official response from WYZE.
Stay tuned.
I’ve never been into fighting with tech at large corps…THIS maybe unorthodox but is a 100% fix. I have a relay driven by a wyze plug. The relay breaks the current to the thermostat. Just go to the plug…on then off and the thermo resets itself. Works every time
Is this something I could install, I have taken to ripping the thing off the wall each time it goes off the wifi which is about once every 2 weeks.
Same thing here. I posted drawings of my arrangement a while back.
@ascully I would be happy to make you a relay box for cost of parts - a small project box and a relay. Less than $20. I’ll happily donate an old lamp cord to power the thing. You simply wire the “C” wire to the box and run another wire back to the “C” terminal on the thermostat. Plug the relay box into a WYZE plug. Config the plug to go off 5 seconds after it comes on. Turn the WYZE plug on, it opens the relay contacts then goes off after 5 seconds and the contacts close again. That reboots the thermostat. Let me know if you are interested.
If the wyze team added a reboot to the actual unit that would fix this wouldn’t it?
Sorry this is a long post, but I have a potential hypothesis inspired by this comment:
This is interesting information. I know there are certain IoT chips that are known to have a conflict with routers’ WMM settings. The original Wyze Bulbs and Wyze plugs had one of the chips that have an issue with WMM, for example. I learned this back when I was using a version of OpenWRT router firmware/software called “Gargoyle Router”…I had a REALLY hard time getting my v1 bulbs and v1 plugs to pair or stay connected. After an insane amount of research I discovered that a lot of IoT devices had a conflict with certain WRT-based routers. Further investigation explained that certain IoT chips conflicted with the WMM setting, and that it was possible to manually turn it off. So I used PuTTY to login to manually alter the router firmware to disable WMM. As soon as I did that, everything suddenly worked.
Then, I later learned there are also some IoT chips that have issues with a couple of other settings, particularly common with WRT-based routers (but not limited to them). Some people have had luck by disabling what is called the “LDPC” sometimes in addition to the WMM, and to “use CCMP only”.
This was apparently something that was happening in a lot of ESP-based devices, at least with some of the older chips. Supposedly ESP fixed the conflicts in some of their newer chips.
Honestly, at this point it’s almost impossible to know if your branded router is using WRT-based firmware or not. Most companies nowadays will take WRT open source firmware as their foundation and then build their own branded custom firmware on top of it. Even Wyze’s own routers are basically WRT-firmware that they modified for their own use. Lots of companies do the same, even big name brands. But regardless of that, we do know that WMM, and LDPC features have had conflicts with some IoT chips, including some ESP chips. And sometimes those are turned on even if your router doesn’t have a setting option to toggle them on/off (I had to use PuTTY to go turn mine off in my old Router for example because the UI didn’t have a WMM setting anywhere, but it was actually turned on anyway). Now I ALWAYS disable WMM, QoS, and LDPC on my routers whenever possible. I’ve seen too many examples of them causing issues with devices, and sadly half the time they aren’t even visible options in the GUI.
My point is that you mentioning WMM reminded me that something like this could very well be related to this issue. It could explain why some people don’t experience this issue at all, while others are experiencing this issue. It’s possible the IoT chip they used doesn’t play well with a setting like WMM or LDPC on certain routers.
It may also explain why Wyze hasn’t made a lot of progress with this issue. The Thermostat can’t see if a router has WMM or LDPC causing it to get kicked off the network. It only knows it is getting kicked off by the router, not why. It’s also possible they can’t program a firmware fix for something like this. If a router kicks a device off, it’s not like the device can FORCE the router to keep it on. It is seriously quite possible that whether the Thermostat always stays on or keeps losing connection may be directly related to which router a person has and whether they have WMM/LDPC on or not. I am not saying it is the router’s fault, or the IoT chips fault. In a way both are at fault. The router works fine for other people with other devices, and the device works fine for other people with other routers. Regardless of fault, this is a pretty plausible explanation for the root of the issue that may need to be explored further.
If I were Wyze, I would have some people with the issue report what exact router models they’re using and try to replicate the issue on those routers and then see if they can fix it. My hypothesis might be wrong, but it’s an interesting avenue to explore.
If anyone gets around to looking through Public FCC files to find out what chip Wyzes uses in the Thermostats, we might be able to search and see if there are any known conflicts with WMM or LDPC or anything like that. That could be useful information.
(Also, I should mention I am not having this problem at all, so I can’t really test/try it out --I originally reported this issue to Wyze and got it added to F-it-F to help the rest of you who are having this issue, and I have been tracking it to keep passing on to them any new information (not because it affects me, but because I like to help people who are having issues like this)-- but as I said, I always disable those settings discussed in this post, so that could be one explanation why my Thermostat stays online all the time while others are having the issue).
Makes a lot of sense to me. I have not had the issue a single time at my cabin in the woods after I turned off WMM.
I turned off WMM in my router at my house and the f
problem occurs less frequently.
I will look at the other settings you mentioned when I can. Maybe that will fix the issue at home.
Only if the thermostat sensed itself offline and initiated a reboot. With the unit offline, the WYZE app can’t get to it to reboot it.
I like your thinking. I agree, this might be a decent workaround that Wyze could implement on the firmware side, even if it turns out to at least partially be related to router settings. Wyze could setup a local rule that if the Thermostat loses connectivity for X amount of time, then force it to restart itself. This should hopefully force it to re-establish the connection with the router from scratch and thus reconnect.
I think that should be a minium solution from Wyze. Maybe make it a setting toggle where a user can select if it does that or not and how long it should wait (after loosing connection) to try restarting. This could actually be a pretty decent resolution if it turns out to be a router setting vs IoT chip incompatibility. At least then it would connect back on it’s own automatically.
Not sure if it’s been discovered and mentioned in the past either by me or someone else. This past week I spent a lot of time at the house where my thermostat is located (normally I am remote). What I discovered is that it rarely went offline or so I thought. At one point I checked it when I was out getting lunch and it was offline. So I left it until I got back to the house to troubleshoot. The second I was within that WIFI network it was back online.
This happened repeatedly and was reproducible simply by disconnecting my phone from WIFI and going to Cell data. What I deduce is that the thermostat isn’t loosing connection to WIFI or the internet, It’s just not logging into the Wyze server to provide connectivity over the internet.
Edit… This is exactly what I reported back in April 2023 so I guess this is just a repost LoL!
Adding to the chorus of voices. Mine worked flawlessly since late 2022 or early 2023 can’t remember.
HOWEVER, when looking into the device info on the device itself it has 0.0.0.0 as an ip. Not 169.254.x.x
Google Mesh states the device is connected and responding to pings.
App version 3.2.7(1)
Thermostat Version 1.2.8
Thermostat offline again. Log 1719848 @WyzeJasonJ
FWIW, I swapped the two thermostats at my primary residence- moved the one from the master bedroom to the hallway and vice versa. The problem followed the device. So, it’s not the location in the house, it’s the device. The other one has not gone offline to my knowledge.
Said it many times before in the forum, but I wish Wyze would own up to the fact that there’s a problem and then dedicate resources to fix it.
Just dreaming.
Wyze couldn’t ask for better Field Testing!
I was thinking that Wyze had taken this Thermostat off the market by now but I checked on Amazon and they claim that they sold over 500 in the last month!
If they weren’t selling any, I could see abandoning it but if they are still selling them, they need to get it fixed!
The logic you describe is pretty common in any firmware/software that operates remotely. This should have been in the very first version.