Hello! I recently got my very small zigbee network up and running, which was running well for a few days, but now all of the sudden my bulbs are both “unavailable.” I’m using the zha integration.
I’m running Home Assistant 0.110.1 on a pi 3b+, with a conbee.
Rebooting. power cycling, removing and re-inserting the stick, have all born no fruit. The logs for core seemed to indicate nothing odd occurred around when the devices first went available (as indicated by when the bulbs were “last seen” in the zigbee configuration panel). Even more weirdly (maybe?) the conbee was last seen a few days before the bulbs were.
Does anyone have any ideas as to how I might fix this? Or what information I should gather? I had a similar issue out of the blue with my network when I was using the deconz community add-on (which I was using reliably for over a year). Once that issue seemed difficult to shake, I decided to burn it out and jump over to the zha integration because I don’t need much fancy and I assume(d) that the zha integration is a little more of a “core” functionality.
Thank you! I did so, and then grabbed the logs from after rebooting. I filtered out lines that didn’t include zha and zigbee because I’m not sure if there’s any sensitive information in there.
I can’t make heads or tails of it… perhaps some kind soul is willing to take a look at the log?
I’m no expert but it seems there are deconz and zha commands in the log, I guess it should be either one and not both. You might try to remove all deconz related things, but this is all advice I can give.
Hoperfully someone with more knowledge will come along.
In looking at the zha debugging guide, it mentions turning on logging for zigpy_deconz.zigbee.application and zigpy_deconz.api so I don’t think that that is the issue.
When trying to add a different device today, I saw this in the adding devices window regarding one of the unvailable devices…
light.bedside_his: polling current state
[0x788f] Delivery error for seq # 0xd1, on endpoint id 11 cluster 0x0006: message send failure
[0x788f] Delivery error for seq # 0xd3, on endpoint id 11 cluster 0x0008: message send failure
[0x788f] Delivery error for seq # 0xd5, on endpoint id 11 cluster 0x0300: message send failure
[0x788f:11:0x0300]: failed to get attributes '['color_temperature', 'current_x', 'current_y', 'color_loop_active']' on 'light_color' cluster: [0x788f:11:0x0300]: Message send failure
I ended up factory resetting my devices and starting from “scratch.” I also put my stick on a 1m usb extension cable, and switched to a “nicer” 3A power supply for my pi. A week later and it’s still going well… fingers crossed. I’ve also enabled all of the logging recommended in the zha debugging guide.
Sorry for ressurecting such an old thread, but @Dolores, how did you factory reset the dongle it self? I get the same strange behavior as you seem to have had and get “message send failure” on some bulbs, which also seem to happen randomly.
I didn’t actually factory reset the stick, just removed the integration (and probably waited a bit) then rebooted the system and re-added integration.
It happened one more time, and then I realized that my wife’s work laptop was near it, and she only uses that sporadically. I moved the stick a few feet from there and haven’t had issues yet. May have been a coincidence but .
Good luck!
.