-
-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
SONOFF zigbee 3.0 usb dongle plus Keeps crashing #12566
Comments
Can you provide the zigbee herdsman log starting from the point where you start z2m until it fails? See https://www.zigbee2mqtt.io/guide/usage/debug.html on how to enable the herdsman debug logging. Note that this is only logged to STDOUT and not to log files. |
|
I've disabled Availability before, but it didn't make a difference. Only that there is a less visual way for me to see that there is an issue. A bit difficult to look back in dmesg... However, I bought a new SONOFF and have moved everything over to it instead. Will see now if it works any better or has the same issues. |
I had the same problem with my Electrolama stick. Check if you have the Zigbee Home Automation integration installed. When I disabled the integration everything worked as it should. |
Nope, same thing happens with the new dongle. It was going strong for the first couple of days, but then last night... ☠️ Also, no I do not have ZHA. Again, the issue does not occur when I use the ConBee II. |
Same here. And some Philips motion sensors get partley disconnected (still shows connection, but it sends no updated sensor data). After hard restart (power off/on) everything is fine until it happens again after some days. Pairing of new devices is not possible as well in that state... |
At the moment for me, it happens once to twice per day. Most of the time, I just restart the add-on and it’s fine. It’s easy to spot because all of the devices become unavailable in HomeAssistant |
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days |
Issue still occurs. Still haven't really discovered why... |
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days |
FYI, I've updated the firmware to 20220726 and it seems to work better, but it still happens. Ie, longer periods between when it breaks. Now it's maybe once a week. |
Hello, Thanks for your feedback |
Here -> https://github.com/Koenkk/Z-Stack-firmware/tree/develop/coordinator/Z-Stack_3.x.0/bin |
@Veldkornet Exact same issue here! I am also running; 20220219 |
@Koenkk is this a known issue and solved with an newer firmware? |
@popy2k14 can you provide the herdsman debug logging starting from the moment you start z2m until the crash? See https://www.zigbee2mqtt.io/guide/usage/debug.html on how to enable the herdsman debug logging. Note that this is only logged to STDOUT and not to log files. |
@Koenkk i have enabled the herdsman logging on my hass.io supervised addon and tried to gather the logs as described in the link. There are now herdsman entries in it. When the issue occured i will grab them an come back here. Did you found an issue in @Veldkornet log above? @Veldkornet is the new beta firmware 20220726 stable for you? |
The new firmware works better, but it still happens unfortunately. |
@Veldkornet thx for the info. @Koenkk it happened again, here is the log file: https://drive.google.com/file/d/1NjKBKp9Qz-TPUm7kYP6pZS8MHRjekf8a/view?usp=sharing |
@popy2k14 the log doesn't stop with a crash, just with:
@Veldkornet do you by any chance of have SONOFF sensors on your network (I mean battery powered sensors; not the coordinator itself) @inertia666 I need herdsman debug logs of the crash, otherwise I cannot help. |
@Koenkk When it help's i can change my "test" and do not do an network scan the comming days and just try to see if the devices are offline. Please let me know how i should proceed with my test. |
I'm happy to report that my USB stick is working perfectly well now. I had misunderstood the instructions and exposed the device to two different Docker containers (Both Home Assistant and Zigbee2MQTT). Presumably, this was causing the crash. Sorry for reporting that I was experiencing crashing. |
@popy2k14 does z2m still answer the network map scan? If yes then z2m it self has no crashed. I do not see the finish of the network scan in your logs. |
@Koenkk yes, it returns an error in red. |
add the following to your config adapter: ezsp ^^ I saw someone post this, and it fixed 4 days of trying to make it work, but now Z2M crashes every 15 minutes. |
Thanks for your help. I'll try this later and come back then |
So now is later :) Finally found another container interfering with z2m. In my case raspberrymatic (Homematic container'd) needs to be switched off before starting z2m. For any reason, it looks like Homematic is taking over the stick. So it's a matter of the starting order. I'll write another docker-compose.yml which will start both container in there respective order. Clear to me now why the container was working on my laptop. No other container there interfering with z2m. @srvoets your idea with adapter: ezsp didn't help. If set, adapter will be accessible but z2m won't start. I removed the entry and now it works like a charme. Thanks for your help anyways. |
Ahhh, I pulled my hair out for days seeing the same exact timeout error over and over again, and saw people mention make sure nothing else is running that could use stick, removed z-wave uninstalled it, had same issue, so I tried it in a docker container in unraid, same issue. only thing that made it work was adapter: ezsp So you are running this adapter without the above 'adapter' option in your serial config? Very interesting that yours works, and mine does not, is yours running stock firmware? I tried for several nights to flash it using the TI tool, python app, remove case holding buttons, never got it into bootloader mode to take the new firmware. I really wish I did though because my zigbee network runs like dogshit now, I might go back to conbee until I can get this flashed and then set it up as a router. My other consideration is migrating Z2m from from my unraid build inside an HA VM to a spare Pi that could be placed more centrally. However, in the original location, conbee worked without issue, I blame this whacky stick. I even unplugged my steelseries headphone\mouse wireless adapter, but no difference. At least it "runs" |
@srvoets FYI my docker-compose.yaml which is working:
and configuration.yaml:
And take care of the starting order. As said in my case Z2M needs to be started first. Then Stick is attached to z2m. Hope that helps. Just had a look to the preview. Looks like format of .yaml's lost to copy paste. Sorry for that |
Hello, I also switched to Sonoff Dongle-P, which ran like a charm for a week. But it stopped working almost yesterday. now its crashing almost immediately after start. Curious thing ist - didnt change anything... Some information (everything the same for working state and crashing state):
dmesg log shows the following:
Any ideas what i could try next? Am a little frustrated right now - migrated from deconz with conbeeII to Z2M with conbeeII (still some problems with lights not responding) and finally to sonoff Dongle-P. Was so happy that finally everything was working as it was supposed to. (including availability) |
Please try |
Thanks for the reply! Installed it right away, but sadly the errors didn't go away. I have the feeling that the problem may be at driver/linux usb side. Will look into it, i really hope to find a solution! |
@klaus1k @srvoets |
Tried running with root privileges, sadly didn't help. So i tried unloading and reloading Then i noticed, that i get an error from
EDIT: apparently adding |
I just joined this thread due to having crashes with the SONOFF ZigBee 3.0 adapter. I've upgraded today to The nice news is that after 8 hours of downtime and then upgrading to |
I've applied |
Nope, already crashed.
|
@Veldkornet can you provide the herdsman debug log starting from the moment you start z2m until it fails? See https://www.zigbee2mqtt.io/guide/usage/debug.html on how to enable the herdsman debug logging. Note that this is only logged to STDOUT and not to log files. |
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days |
The |
Right after that, I went full nuclear ☢️ Completely removed anything Z2M related, reset all devices, started from scratch. It hasn't gone down once since then, more than a month already. I've even started to revive some old sensors and add them back to the network. Going strong 💪🏼 I was hesitant to report back at first, in case I jinx it 😄 |
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days |
@Koenkk Back in business. I have now again an Sonoff Zigbee 3.0 USB Dongle Plus with latest 20221226 firmware. And yes, also with the new firmware, it crashes! PS.: after an addon restart, all is working again. the container/addon is stopped but the Zigbee2MQTT frontend is reachable. It just shows no lines on the network scan, and no command (on/off) is working nor data is received from the devices. Here is the herdsman log:
It crashed (doesnt work) around: 18:54:12 BUT this time i added the devices in the following order:
It worked without an crash ~6h, but after i added the last device (SML002) it crashed twice in short time (minutes). Maybe there is an issue with the hue motion sensor? Ps.: The sensor is working without an issue on deconz! So it's NOT defect! |
Found the following in my dmesg around that time it failed:
That seemed to be an long ongoing kernel bug which is not solved up until now. raspberrypi/linux#2406 Ps.: Running an Odroid N2+ The issue seems to affect various manufactures of USB uarts, see the issues linkes above. What seems to help is using an active powered usb hub in between, not because of the high power draw of the usb-uart devices, rather to wokaround the issue. If the issue pops up again, i will throw an unpowered usb hub in between and see if the issue is gone. What's a little bit wierd is, that my conbee II with deconz addon running good for months. Does zigbee2mqtt some retries of this sort? |
Today it crashed again, in dmesg are the exact same messages. Will migrate to an x86 hardware and see if the issue is gone. Sorry for spamming this thread. But maybe this info still help someone, with the same issues. Will post an update after the migration with the exact same config (HA full backup) when it's done. |
Running the new x86 hardware with an HA backup for an day and so far so good, no crash! |
After 2 weeks, Z2M on x86 with sonoff dongle p is rock solid! |
You saved my day @adfx , big thanks! |
This might be a "stupid" thing to blurt into this conversation here, but everything was fine and dandy for the past months without any hiccups or anything (apart from some huge delayed response times from devices but that's beyond the point). Just as a "this is my experience" kind of info dump here :) |
What happened?
As a continuation of #11216, every so often all devices become unavailable and I get messages like the following:
I currently run both MQTT and Z2M on HomeAssistant OS on a RaspberryPi 4 with a ARGON one M.2. The Sonoff adapter is plugged into a Nedis Powered USB HUB, which is plugged into one of the usb 2.0 ports of the RPi.
The only items plugged into the hub are the SONOFF and an Aeotec ZStick.
This issue does not occur with the ConBee II.
I’ve been through the troubleshooting topic listed here; and I’ve added
usbcore.autosuspend=-1
, but this hasn’t seem to make any difference.It can crash every few hours or every 1-2 days, I don’t find a clear pattern.
What did you expect to happen?
Everything should stay running and the coordinator should not crash.
How to reproduce it (minimal and precise)
unclear, it just crashes on it’s own.
Zigbee2MQTT version
1.25.1
Adapter firmware version
20220507
Adapter
SONOFF Zigbee USB Dongle Plus (zStack3x0)
Debug log
No response
The text was updated successfully, but these errors were encountered: