Deconz-rest-plugin: Random lights can not be controlled

[Lindansaren]

Describe the bug

Random lights can not be controlled from time to time. Pressing 0 in VNC on the node gets it o work and respond properly again.
Blinking red light on node in VNC when it happens. Have about 80 nodes in total. About half of them powered by mains and spread out evenly through the house.

Steps to reproduce the behavior

Lights should react to commands. =)

Screenshots

Environment

• Host system: Raspberry Pi4
• Running method: Home Assistent deCONZ Add-on
• Firmware version: 26660700
• deCONZ version: 2.05.84
• Device: ConBee II
• Do you use an USB extension cable: Yes
• Is there any other USB or serial devices connected to the host system? If so: Zwave.me stick

deCONZ Logs

01:02:19:924 delay sending request 227 dt 8 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:023 delay sending request 227 dt 8 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:123 delay sending request 227 dt 8 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:223 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:324 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:423 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:524 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:623 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:724 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:824 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:20:924 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:023 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:124 delay sending request 227 dt 9 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:224 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:324 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:424 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:523 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:624 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:751 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:825 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:21:923 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:023 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:123 delay sending request 227 dt 10 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:267 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:323 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:423 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:523 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:623 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:724 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:823 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:22:924 delay sending request 227 dt 11 ms to 0x90FD9FFFFE33A6D9, ep: 0x01 cluster: 0x0006 onAir: 2
01:02:23:013 0x90FD9FFFFE33A6D9 error APSDE-DATA.confirm: 0xA7 on task

Additional context

[Mimiix]

Can you update to the latest version? There are some fixes in there.

Also, are you able to provide a screenshot of your deCONZ map?

[Markkuuss]

@Lindansaren: If the lights do not react to your command, try switching the corresponding group. What happens then? Do the lights still not react or do they then switch on/off with the group?

[Lindansaren]

I do not have groups set up in deconz only in HA. Do you mean groups in Deconz?

[Lindansaren]

[Mimiix]

Did you update to the latest version?

[Lindansaren]

Yes, I have. Still issues.

[Lindansaren]

Pressing a 0 on the node gets it back to life.

[DerOetzi]

Same issue here. Having about 60 nodes at the moment. There are 3 IKEA TRADFRI bulbs and 1 osram plug in different rooms of the house, which not reacting randomly on any REST call by phoscon or openHAB. Yesterday I had not the time to directly go to deconz. After about a hour the bulb which is triggered together with 5 other bulbs by a motion sensor via openHAB starts reacting again without any interaction from my site.

Version: 2.05.88
Firmware: 26660700

[Lindansaren]

@DerOetzi I have experienced the same. Leave the light for an hour or so also gets it to react. According to my mesh the node have several "lines" connected. Should not be an issue.

[Lindansaren]

My wife is not impressed when her bedside lamp decides to not communicate anymore. Or when the nightlamp in the bathroom connected to a motion sensor does not turn on. =(

[DerOetzi]

@Lindansaren are those always the same nodes not reacting like on my side?

[Lindansaren]

I have done some more investigations today and noted that it's boiling down to maybe 10 different nodes. 8 IKEA (all versions including outlet) and 2 Philips Hue (E14)

[SwoopX]

@Lindansaren You seem to have too much in the air simultaneously. With all those nodes, you sure you're using unicasts (multiple lights going on only one by one e.g.)? Using zigbee groups and therefore groupcasts should ease such situations and also be resilient to routing issues.

I mostly only notice after a couple of days that my smart home system is down (middleware seems to have a memory leak somewhere) while primarily using groupcasts.

[Lindansaren]

@SwoopX The problem is that this is occuring not only when sending multiple commands. This is also an issue when trying to reach individual nodes manually from HA. As stated before. Going in to VNC and pressing 0 on the node make it controllable again.

[Markkuuss]

I do not have groups set up in deconz only in HA. Do you mean groups in Deconz?

Yes, I mean groups in Deconz. Add the affected light in Deconz to a group and try to switch the group via API or WebApp.

I have a similar problem with a Ubisys S1-R. It doesn't react anymore when I control it directly. But it can always be switched via the group. Can you please check if you have the same behavior.