Hi everyone,
Has anyone else experienced issues with newer Aqara FP300 sensors in Homey via Zigbee?
My setup:
Homey Pro 2019 (no Matter support)
Aqara app in Homey
FP300 paired directly via Zigbee
What makes this confusing is that one FP300 installed about a week ago still works perfectly:
presence detection
lux
temperature
humidity
timestamps updating correctly
However, multiple newer FP300 units now behave like this:
pairing succeeds normally
initial values for “Motion Alarm” and “Presence Alarm” appear random/inconsistent immediately after pairing
those states never change afterwards, regardless of movement or occupancy
temperature, humidity and lux initially show “Last value unknown”
after a few minutes, Homey starts showing “56 years ago” for those sensors
the devices then appear effectively non-reporting
Troubleshooting already performed:
Remove Device in Homey
factory reset of FP300
direct Zigbee re-pairing
pairing without Aqara app involvement
repeated tests with multiple FP300 units
Result is always the same on the newer units.
Additional observations:
It is unfortunately not possible to check the FP300 firmware version once connected directly to Homey.
I do not want to experiment with the one currently working FP300, since it is stable and functioning correctly.
Resetting the entire Zigbee network in Homey is also not an attractive troubleshooting step, since it would require re-pairing a large number of already stable Zigbee devices.
At this point I suspect either:
a recent FP300 firmware incompatibility
or a regression between newer FP300 firmware and the current Homey Aqara driver.
The older FP300 still functioning correctly suggests the Zigbee network itself is healthy.
Has anyone else seen this behavior or found a workaround?
Any help would be greatly appreciated. Thanks!
Decive settings show a type description like “lumi.blah.blah”
Check the 1st post of the Aqara topic to see if it’s mentioned and thus supported.
China targeted devices have different type descriptions.
Good morning, I have the same issue since a couple of days. Both my FP300 stay in the same modus as presence detected. Tried rebooting homey, preparing. Nothing works
Is your Homey having latest stable fw release 13.2.0? If yes, then updating to experimental release 13.2.1-rc.4 will help, at least in my case. Lot of problems especially with Aqara zigbee devices with 13.2.0, but workin as normally again with 13.2.1-rc.4.
I know what consequences there might be with experimental updates but now it seems that even “stable” releases can break Homey, like now happened.
And there is always possibility to go back to stable releases, just unselect experimental updates from the dev tools, and whenever there is next official stable release, you are back to official firmwares again, no more experimental updates.
At least I have disabled “experimental updates” from dev tools, back in time, and after that I received stable updates only. Just couple of days ago I make this selection again, to get rid of zigbee issues.
Adding a data point on a firmware version I haven’t seen reported in this thread yet, and I think I’ve narrowed it down to the platform side.
Setup:
Homey Pro (Early 2023)
Homey firmware 13.2.1 (stable, not the rc builds)
Aqara app, 4x FP300 paired directly via Zigbee (no hub)
~4 months of near-flawless operation before this started
Symptom (began ~3 weeks ago, no change on my end):
3 of 4 FP300s fire false motion-alarm events with no one present, roughly every 20-30 minutes
The motion alarm gets stuck ON - presence correctly clears, but motion never resets
10+ second delays before an alarm is raised
One unit hangs on motion within hours of every re-pair regardless of what I try
Why I believe this is platform-side, not the device:
Two days ago I reflashed all four FP300s to the latest firmware via the Aqara app over Bluetooth, then did a fresh re-pair to Homey. The problem is unchanged after two days. Combined with the steps below, that rules out the device itself.
Already ruled out:
Latest FP300 firmware (reflashed all 4) + fresh re-pair
Factory resets, AI retraining
Full Zigbee mesh rebuild + added repeaters near the units
No HVAC correlation; no other Zigbee devices on the mesh affected
Homey hub not memory/CPU constrained
I saw the reports that 13.2.0 broke Aqara Zigbee handling and that 13.2.1-rc.4 fixed it for several people. My situation looks like the same regression, but I’m on 13.2.1 stable and still seeing it.
The one question that decides my next step:
Does the fix from 13.2.1-rc.4 actually carry into 13.2.1 stable, or not? If it does, I’m chasing a different cause. If it doesn’t, I need to know whether the only fix is going experimental, or whether a future stable will roll it forward.
Specifically:
Anyone else on 13.2.1 stable (not rc) still seeing FP300 motion alarms stick ON while presence clears?
For those whose FP300s are stable - what Homey firmware are you on, stable or experimental?
Happy to provide logs, screenshots, or video. Thanks.
Thanks, Peter. That’s the confusing part - I’m on 13.2.1 stable, so per the changelog the fix should be in, but I’m still getting the exact stuck-motion behavior. Even since writing this, I’m now seeing it on my 4th FP300, with both motion and presence getting stuck. This is after reflashing all 4 FP300s to the latest firmware and re-pairing two days ago, so the device side is ruled out.
So whatever the 13.2.1 fix addressed, it didn’t cover this specific symptom. Has anyone else on 13.2.1 stable seen the motion alarm specifically get stuck (as opposed to devices dropping off entirely)? Trying to figure out if this is a known residual issue or something new worth reporting to Athom.
Apologies for the frustration coming through - for four months I’d only see a very rare hiccup, but now these FP300s seem to have a mind of their own. I have zero issues with any other devices on my Homey, and specifically zero issues with any other Zigbee devices. It’s only the FP300s.
All I know about Aqara is they tend to make it hard for 3rd party zigbee controllers, like Homey, by not following general zigbee 3.0 guidelines (manufacturers are allowed to publish their own flavour unfortunately). Why now only the FP300s are acting up…
Interestingly, within the last hour or so, everything has started to stabilize. I still need to give it some time before I can say for certain that the issues are resolved.
What I still don’t understand is why the problems began about three weeks ago and then, within the last week, suddenly became completely out of control.
My 13.2.1 update installed last night, but I continued having major issues all day with devices dropping off and motion alarms hanging. I’m hoping the AI just needs some time to fully adapt and finally deliver reliable results.
I’ll write back tomorrow and let you know how it’s going.
Quick update, Peter - since the firmware reflash on the FP300s, they have been noticeably more responsive overall. That said, I’m still getting the occasional hang, and now it’s both motion and presence freezing together rather than just motion. I’ll keep watching and report back if anything shifts.
For context on why I’m pushing on this: a hang once or twice a month on battery sensors, fixable by just walking back into the room, was completely acceptable. Having them hang indefinitely and requiring a sensor reset to recover is not.
I am having the same issues with my FP300 in Home Assistant. The Presence sensing is working as expected but the PIR sensing goes off constantly. It will detect motion and then a minute later clear. The device worked great when I first got it.
Jason - this is really useful, thank you for posting. Your symptom pattern (constant PIR false triggers with ~1-minute clear cycles, started after working fine initially) matches mine almost exactly, and you being on Home Assistant rather than Homey is significant. It means this isn’t a Homey integration issue - it’s something on the device side that’s hitting users across multiple platforms.
There’s also a recent thread on the Aqara forum where users on other ecosystems (Apple Home, Home Assistant) are reporting similar absence-detection and motion behavior issues starting around the same time, which lines up with a recent FP300 firmware push. So we may be looking at a cross-platform firmware regression rather than anything platform-specific.
A few questions to compare notes:
What FP300 Zigbee firmware version are you on? Aqara support told me the latest is 0.0.0_6542 and described it as addressing radar/PIR false-alarm behavior - so the question is whether we’re both already on that build and seeing residual issues, or on an older build that the fix would resolve.
Have you tried reflashing the firmware via the Aqara app over Bluetooth? I did all 4 of mine ~3 days ago and saw partial improvement but the hangs still come back.
Roughly when did your problems start? Mine began about 3 weeks ago after ~4 months of flawless operation.
Are you running the FP300 in Zigbee or Thread mode in Home Assistant?
