Is it though? I cannot find much in the Hubitat Community or even on the web about this being a widely known issue with these sensors, generally, or with Hubitat, specifically.
I'm using the Kossev driver V2.0.1.
I'm using 2A power supplies, mine did not come with any. What were you using for power, and did you try swapping them out?
I don't like adding extra hardware for "fixing" another hardware problems. Any RM rule will be somewhat acceptable if it works. However @danabw already tried this workaround and it was not successful. If I am not mistaken he is back with power recycling 'fix".
Yeah power cycling is not a fix. Its one thing if it happen randomly once a year, but not at any frequency.
I am sorry, I am too lazy to dig through the threads but definitely it was few related involving myself, @kkossev and @danabw . I am experienced EE, had installed 4 sensors (2 and 2 bought at a different time) and all 4 exhibit the same failure. To my eyes it was enough evidence for replacing these sensors. But who knows, maybe it was huge batch of failing sensors or possible a fix with different firmware. My hope at some point it could be a firmware fix. That is why all 4 of these toys are waiting in a drawer for the potential fix.
A similar approach was already tried here without success :
I am curious if the issue happens with the device connected directly to the hub and not through a neighboring ZigBee repeater.
Most of mine are showing a direct connection to the hub, except for one, which is showing paths through two repeaters. Most of the sensors are acting as repeaters for other devices as well. So I have a mix of direct, through repeaters, and not all are acting as repeaters for others. One is acting as a repeater and connected through a repeater.
I'm pretty sure that I have gone more two months without rebooting the hub since I got the first three sensors, as I don't do every hub update when it comes out. Having never seen one go unresponsive on its own, maybe I've just rebooted before they reach the unresponsive point?
I'm not sure if that explains sensors that fail to connect after a reboot, were they in a state where they were ready to go unresponsive, and the reboot kicked it in? Really hard to say. It is something I just check after a reboot now.
Might be worth noting that my issue may have been resolved with 2.3.9.200 as I have had no issues for the last several reboots with any of them not reconnecting.
The initial three are LimpTech branded:
|Manufacturer|_TZ3218_awarhusb|
|Model|TS0225|
The two I bought a month ago are Moes branded:
|Manufacturer|_TZ3218_t9ynfz4x|
|Model|TS0225|
I have four of these, two Linptech and two Moes, Aliexpress and Amazon, same thing. Two had no drop out history, the other two were a random two to three weeks. I tossed a cheap Kasa plug in front of all of them and selected twice a week for a very brief middle of the night off/on event twice a week for no carefully determined reason and have not experienced a dropout since. This was almost a year ago.
Same except used the SonOff EweLink plugs and power cycle once a week at 3am or so, and never had any problems since.
These things function great for me in every way except the lockups, and they have very good WAF because my wife likes how they look. A rare victory. 
So I decided to put up with the need to power cycle them.
Maybe Linptech will figure this out at some point and offer a FW update (assuming it would be fixed via FW), or someone will figure out a way to "bump" them via SW, but at this point I don't really think I'll end up swapping the Linptech out for anything else.
They also repeat a lot for me, and I have a mixture of them connecting directly to the hub, and connecting through a repeater. I have them on a C7.
Lol, the only time you ever find my eyes open at that early would be when one of my dogs gets me up to go out for a pee. I have learned that when she says she has to pee she is definitely not kidding. She sleeps in our bed. 'nuff said. 
I still can't understand whether the Linptech sensor becomes 100% unresponsive, or it is just the 'presence' ( motion active) state that locks up ?
When my Linptech(s) became unresponsive (usually stuck in "active" state) they still reported LUX values. At least Zigbee communication still was active and intact. Only "motion" reporting was gone.
I'm was running a SONOFF SNZB-06P with your driver and was having this very issue. It was just the presence that was getting stuck. If I hit refresh, it would immediately go to inactive. I switched the driver to the Hubitat recommended "Sunricher MW Motion Sensor" driver and it hasn't happened in a couple of days. I prefer your driver for it's settings. I'm going to switch back today to see if that was even had anything to do with it.
What do you have the settings on for detection sensitivity and range and the room size? I have Linptech in two medium size rooms and they never get stuck on active. I have a Moe's version in a small 6x9 room and it stuck several times. All 3 sensors were set to the max on settings. I ruled out detection through a wall which led me to believe it could be the settings I had on the sensor in relation to the room size and the metal and glass objects in it. I changed all detection and distance settings to midrange and the sensor performs as it should and has not gotten stuck on active since. This could be purely coincidental but I wanted to share if it may help you in any way.
I replaced all Linptech sensors with Merros MS600 because every 6-8 weeks Linptech got stuck in "active" state. The "fix" is a recycling power to the Linptech sensors. In my case this is/was not acceptable. Now Linptechs are waiting for potential fix (if any).
I have not had an issue with either of my Linptechs and they have been in service for almost a year. The Moe's has been good since the setting change.
I am glad to hear Linptech sensors are working in yours case.
As well as Aqura FP2 are also working for somebody.
In my case (and unfortunately few other users having the same issues) 4 Linptech was failing the same way (stuck in "active") and 2 FP2 always was leaving ghosts regardless what I tried. This is absolutely not acceptable behavior.
I would agree with you. I would also think they are a bad batch of sensors but mine was ordered at various times so they had to come from different lots. I don't know how to pin point it but I truly believe it is linked to the environment of the room and the settings but I have no way to verify it and prove it out.
All my sensors (Linptech and FP2) was ordered at a very different time but all from Amazon. So, in theory they could be from the same (bad) batch but this is hard to believe. Talking about the environment all they are sensitive to spinning (ceiling) fans, reflections from the metal surfaces such as refrigerator doors and can see through the wall. But all the above does not explain why they are consistently failing over pretty much the same time frame and regardless where they are installed (small bathroom or large living room).
Download the Hubitat app
