Philips Hue Motion sensors disconnect from the network

[ccorderod]

Bug Report

What happened

I have 10+ philips hue motion sensors paired to my network. I move them all from a philips hue bridge (along with some light bulbs) where them have been working for a while without any problem. Some of the motion sensors disconnect from time to time from the network now; sometimes they reconnect without doing anything, sometimes not. This happen even with a hue motion sensor very close to the coordinator. If they don't, I have to force remove them from the network and pair them again. The network support more than 100 devices and those are the only devices with this behaviour.

What did you expect to happen

Once a hue motion sensor connect, it should keep connected.

How to reproduce it (minimal and precise)

Difficult to reproduce, as the disconnections are random.

Debug Info

philips hue motion sensors with latest firmware version from Philips zigbee2mqtt version: 1.8.0 latest dev CC253X firmware version: P-2 coordinator, latest firmware 2 CC2530+CC2591 on the network running latest routing firmware

[Koenkk]

I'm having the same problem (https://github.com/Koenkk/zigbee2mqtt/issues/2518#issuecomment-569132387), currently waiting for the problem to manifest itself again, then I can start looking into it.

[WouterJN]

Same issue here with Hue Switches and Motion Sensor.

[papanirual]

Same issue here with Hue outdoor motion sensors. At the moment I am still unable to repair two of them. I removed them from the database of zigbee2mqtt and tried to repair them, but the only thing happening is that there is a line in the log that the device announced itself, without the usual pairing procedure.

Looking forward to any fix - many thanks in advance!

[Koenkk]

Update from my side, right before it stops working it seems that the devices leaves the network:

debug 2020-01-11 10:30:40: Device 'veranda_occupancy_sensor' announced itself
warn  2020-01-11 10:30:40: Device '0x001788010644f56d' left the network

Then it joins again:

info  2020-01-11 10:30:40: Device 'veranda_occupancy_sensor' joined
info  2020-01-11 10:30:40: Starting interview of 'veranda_occupancy_sensor'

But this fails:

error 2020-01-11 10:32:01: Failed to interview 'veranda_occupancy_sensor', device has not successfully been paired

Going to dive further into it.

Question to all: what coordinators and coordinator firmware do you use?

[ccorderod]

Coordinator: CC1352P_2 firmware: 20191106 Most of my hue motion sensors are able to join again, but leave randomly again. @Koenkk : is there is any way to log with DEBUG=* npm start but only logging one sensor? My network is too big to run zb2m with Debug enabled as it create enormous log files :-(.

[Koenkk]

Thats not possible, but setting it to DEBUG=zigbee-herdsman:controller* should give the required info.

[hi1000]

CC253X firmware version: 20190608 Philips Hue (9290012607) firmware version: 6.1.0.18912 and 6.1.1.27575 Bugs: After reboot the coordinator, cannot receive Philips motion sensor's message

Log File: 2020-01-13.15-39-50/log.txt:1/13/2020, 3:41:50 PM - warn: Failed to configure r_MotionD1 (0x0017880104b78b01) ('Error: Timed out after 10000 ms') (attempt #1) 2020-01-13.15-39-50/log.txt:1/13/2020, 3:41:50 PM - warn: This can be ignored if the device is working properly

[papanirual]

CC26X2R1_20191106, Coordinator firmware version: '{"type":"zStack3x0","meta":{"transportrev":2,"product":1,"majorrel":2,"minorrel":7,"maintrel":1,"revision":20191106}}'

zigbee2mqtt version 1.8.0-dev (commit #f04e01f)

Koen Kanters schrieb am 11.01.2020 um 22:41:

Update from my side, right before it stops working it seems that the devices leaves the network:

|debug 2020-01-11 10:30:40: Device 'veranda_occupancy_sensor' announced itself warn 2020-01-11 10:30:40: Device '0x001788010644f56d' left the network |

Then it joins again:

|info 2020-01-11 10:30:40: Device 'veranda_occupancy_sensor' joined info 2020-01-11 10:30:40: Starting interview of 'veranda_occupancy_sensor' |

But this fails:

|error 2020-01-11 10:32:01: Failed to interview 'veranda_occupancy_sensor', device has not successfully been paired |

Going to dive further into it.

Question to all: what coordinators and coordinator firmware do you use?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Koenkk/zigbee2mqtt/issues/2693?email_source=notifications&email_token=ALFQYXOUTQSGBPNQMEAT5B3Q5I4J7A5CNFSM4KEFQ5EKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIWLXII#issuecomment-573356961, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALFQYXIT3BSEEL4US2Q4ULDQ5I4J7ANCNFSM4KEFQ5EA.

[sti0]

Did you have some Osram plugs in your network? In https://github.com/Koenkk/zigbee2mqtt/issues/1474 it's described that Osram plugs prevent a re-join of some devices (maily Hue motion sensors). By myself I got two Osram plugs in the network and Hue motion and Hue dimmer are randomly disconnected. After I removed the plugs from the network it seems to be stable.

[GJT]

CC1352P-2 Coordinator firmware version: '{"type":"zStack3x0","meta": {"transportrev":2,"product":1,"majorrel":2,"minorrel":7,"maintrel":1,"revision":20191106}}'

same problem, no Osram devices.

[emilklindt]

I was experiencing this same problem with the Hue Dimmer Switches, as described here.

A week ago I tried changing the channel of my coordinator to use Zigbee channel 25. After searching for interfering WiFi access points near my apartment, I found that the default channel was quite populated.

So it seems to be an issue with interference, from my best guess. Can anyone verify this?

[Koenkk]

Mine dropped off again today, pushed a possible fix in the latest dev branch, fingers crossed 🤞

[peterforeman]

Running on latest (not -dev), saw this:

[...]
zigbee2mqtt:info  2020-01-24 18:18:40: MQTT publish: topic 'zigbee2mqtt-beneden/motion_tuin', payload '{"battery":96,"linkquality":33,"occupancy":false,"temperature":3.34,"illuminance":3242,"device":{"friendlyName":"motion_tuin","model":"9290019758","ieeeAddr":"0x001788010644892e","networkAddress":43045,"type":"EndDevice","manufacturerID":4107,"manufacturerName":"Philips","powerSource":"Battery","applicationVersion":2,"stackVersion":1,"zclVersion":1,"hardwareVersion":1,"dateCode":"20180828","softwareBuildID":"6.1.0.25261"}}'
zigbee2mqtt:info  2020-01-24 18:19:04: Device 'motion_tuin' joined
zigbee2mqtt:info  2020-01-24 18:19:04: MQTT publish: topic 'zigbee2mqtt-beneden/bridge/log', payload '{"type":"device_connected","message":{"friendly_name":"motion_tuin"}}'
zigbee2mqtt:info  2020-01-24 18:19:04: Starting interview of 'motion_tuin'
zigbee2mqtt:info  2020-01-24 18:19:04: MQTT publish: topic 'zigbee2mqtt-beneden/bridge/log', payload '{"type":"pairing","message":"interview_started","meta":{"friendly_name":"motion_tuin"}}'
zigbee2mqtt:error 2020-01-24 18:20:24: Failed to interview 'motion_tuin', device has not successfully been paired
zigbee2mqtt:info  2020-01-24 18:20:24: MQTT publish: topic 'zigbee2mqtt-beneden/bridge/log', payload '{"type":"pairing","message":"interview_failed","meta":{"friendly_name":"motion_tuin"}}'

Updated to latest-dev now and will pair again. Let's see how long it will last this time ;-)

[hi1000]

I setup three zigbee coordinators in the same big room. They are use the different channels and different keys but the same pan_id and ext_pan_id.

[WouterJN]

Mine dropped off again today, pushed a possible fix in the latest dev branch, fingers crossed 🤞

When will this be released in the main branch?

[Koenkk]

@WouterJN in Zigbee2mqtt 1.10 scheduled for Wednesday/Thursday.

[Koenkk]

Mine hasn't dropped of since 2 weeks so things seem to have improved.

[pepp86]

I have reconnected my two philips hue motion sensor yesterday with latest dev and they are lost again during the evening.

Sorry - my fault. My HA System did not show the right value.

[ccorderod]

@Koenkk Hue motion sensors started to drop off the network again yesterday after upgrading to 1.10.0-dev yesterday. Unfortunatly I can't tell wich version I was running previously (it was dev, but I do not know the build #). Will restart z2m with DEBUG=zigbee-herdsman:controller* trying to catch a drop.

[ccorderod]

@Koenkk Not sure if this will help you in any way, but here you go an extract: 2020-02-21T08:20:30.487Z zigbee-herdsman:controller:log Starting with options '{"network":{"networkKeyDistribute":false,"networkKey":[x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,x,],"panID":7100,"extendedPanID":[y,y,y,y,y,y,y,y,y,y],"channelList":[20]},"serialPort":{"baudRate":115200,"rtscts":true,"path":"/dev/ttyACM0"},"databasePath":"/opt/zigbee2mqtt/data/database.db","databaseBackupPath":"/opt/zigbee2mqtt/data/database.db.backup","backupPath":"/opt/zigbee2mqtt/data/coordinator_backup.json"}' 2020-02-21T08:20:31.884Z zigbee-herdsman:controller:log Disable joining 2020-02-21T08:21:57.627Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1030(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T08:21:57.671Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T10:29:48.443Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T10:29:51.724Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T10:29:51.725Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T10:29:51.725Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T10:29:51.736Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T10:29:58.155Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1030(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T10:34:05.249Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T10:34:08.874Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T10:39:04.785Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) Please note this is only an extract. there are hundreds of records with DefaultResponse, as well as many device announcement records. The hue motion sensor also joined and was accepted a few more times than what I have included on the post. Let me know if you need anything else on my side. By the way, I have moved the hue motion sensors to a hue bridge but the one showed on the log in the meantime as my setup is mostly useless without them. Thank you.

[ccorderod]

@Koenkk Apologies for so much post, but I just got a drop :-). Hue motion sensor keeps announcing itself, join, is accepted, not interviewed, join again ... 2020-02-21T11:01:59.354Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1030(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:01:59.443Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:02:04.384Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:02:07.651Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:02:09.341Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:02:14.282Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:02:19.252Z zigbee-herdsman:controller:endpoint DefaultResponse 0x0017880104b4e792/2 1024(10, {"timeout":10000,"defaultResponseTimeout":15000,"manufacturerCode":null,"disableDefaultResponse":true}) 2020-02-21T11:02:22.505Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:03:22.900Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:04:22.306Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:05:56.998Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:05:56.999Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:05:57.000Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:05:57.000Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:07:01.082Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:07:01.082Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:07:01.082Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:07:01.089Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:17:34.210Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:22:00.076Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:27:21.481Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:27:21.481Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:27:21.482Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:27:21.483Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:27:21.483Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:27:21.483Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:27:21.484Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:35:47.820Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:37:30.182Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:50:35.146Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:52:46.654Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:52:46.655Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:52:46.655Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:52:47.392Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:53:09.056Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T11:58:37.810Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' joined 2020-02-21T11:58:37.810Z zigbee-herdsman:controller:log Device '0x0017880104b4e792' accepted by handler 2020-02-21T11:58:37.811Z zigbee-herdsman:controller:log Not interviewing '0x0017880104b4e792', completed 'true', in progress 'false' 2020-02-21T11:58:39.007Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792' 2020-02-21T12:03:54.281Z zigbee-herdsman:controller:log Device announce '0x0017880104b4e792'

[GJT]

same for me after upgrade, when i try to reconfigure I get this:

zigbee2mqtt:error 2020-02-21 19:12:32: Failed to configure 'zb_sensor_kueche', attempt 1 (Error: Bind 0x001788010213f153/2 genPowerCfg from '0x00124b001ca61201/1' failed (Error: AREQ - ZDO - bindRsp after 10000ms) at Endpoint. (/zigbee2mqtt-1.10.0/node_modules/zigbee-herdsman/dist/controller/model/endpoint.js:235:23) at Generator.throw () at rejected (/zigbee2mqtt-1.10.0/node_modules/zigbee-herdsman/dist/controller/model/endpoint.js:6:65))

[Koenkk]

I've tweaked the route discovery a bit therefore it could be a one-time thing (and afterwards it should improve).

Could you try repairing it once for now and monitor if it drops off again?

[ccorderod]

Hi @Koenkk I upgraded to latest dev a few minutes ago. The motion sensor has been working just a few minutes and dropped again. The funny thing is come to live again from time to time, and drop again :-(.

[Koenkk]

@ccorderod can you update to the latest dev branch and sniff the traffic when this happens? https://www.zigbee2mqtt.io/how_tos/how_to_sniff_zigbee_traffic.html

[Koenkk]

For CC26xR1 and CC1352-P2 users, I believe this is an issue with the firmware itself, I've asked TI for support: https://e2e.ti.com/support/wireless-connectivity/zigbee-and-thread/f/158/t/883629

[ccorderod]

Thank you @Koenkk. I use a cc1352-P2 coordinator. Let's see TI answer to your ticket. Thank you for the update.

[Koenkk]

TI has provided a possible fix, which I'm going to test now. ~Firmware: https://drive.google.com/open?id=1Ls8H3R8n2nUxyh9fq-l7hdwdPWg-UD5h~

https://github.com/Koenkk/Z-Stack-firmware/tree/develop/coordinator/Z-Stack_3.x.0/bin

[GJT]

CC1352P_2 with CC1352P_2_20200229.zip firmware and zigbee2mqtt-1.11.0

Sadly still experiencing hue sensor drop outs. Had 2 out 6 of sensors go dark, one never came back and one after about 30-60 minutes two times.

I don't have a stick to sniff traffic, but got one ordered.

[point-4ward]

I upgraded to the new firmware 6 days ago as per my other issue (linked immediately above this post).

Everything has been perfect until a few hours ago.

My kitchen sensor went offline at 2008 hours, came back of its own accord 62 minutes later at 2110. The last reported link quality was 102 when it dropped off, which is slightly higher than average for that sensor.

Then my bathroom sensor went offline at 2107. Link quality was 96 (which is bang on average for that sensor) when it dropped off. It is now 2346 hours here and the sensor is still offline.

I'm noting the times specifically because you'll note that they're almost 'in order', kitchen goes off, then kitchen comes back and bathroom goes off within minutes of each other (could have actually been the exact same time and it's just the next update times).

Hope this helps diagnose the issue.

I'm off to bed now, if the bathroom one isn't back online in the morning I'll let you know whether a quick press of the button works or whether a full re-pair is required.

No obvious issues in the logs for any of the relevant time points, but I'm not currently logging debug or herdsman so :man_shrugging:

Ooooh, should mention. I don't have any osram bulbs but I do have 3 innr bulbs. All 3 are in the kitchen, and the bathroom is above the kitchen so it's possible that both affected motion sensors are using an innr bulb as a router - I'll run a map in the morning and see.