search

zigpy / zha-device-handlers

ZHA device handlers bridge the functionality gap created when manufacturers deviate from the ZCL specification, handling deviations and exceptions by parsing custom messages to and from Zigbee devices.
Apache License 2.0
777 stars 703 forks source link

[BUG] TERNCY-SD01 paired but buttons not working #864

Closed jfisbein closed 2 years ago

jfisbein commented 3 years ago

Describe the bug After adding TERNCY-SD01 device to home assistant,it does not fire any zha event when pushed/rotated.

To Reproduce Steps to reproduce the behavior:

  1. Add TERNCY-SD01 devide to home assistant using ZHA integration (v 2021.4.6)
  2. Device is recognised by home assistant correctly
  3. Go to DEveloper Toosl -> Events
  4. in "Listen to events" fill zha_event and click "start listening"
  5. Click on the Terncy SD01 butoon an rotate it.
  6. No event is shown.

Expected behavior Expected events to get fired when pushing an rotating the Terncy SD01

Screenshots Device correctly configured image

Additional context I tried to remove the device and adding it several times. I also updated the firmware through OTA. I has not worked with the stock firmware nor the updated firmware.

ZHA Pairing log zigpy.log blakadder info

kukachik commented 3 years ago

Which stick do you use? I have this problem on ti cc2538 and on cc2530. But terncy work perfectly on itead erf32 stick. In my case, terncy dont work in z2m too. As i see, this is compatibility problem with ti zigbee stack.

MattWestb commented 3 years ago

In the log its one zigpy_znp coordinator.

kukachik commented 3 years ago

You can try low level serial log, maybe it will be not empty.

jfisbein commented 3 years ago

I use a ZZH zigbee stick.

zeon-ua commented 3 years ago

I do have exactly the same behavior. I use Conbee2 dongle which works perfectly with my other ZigBee devices

bbrandal commented 3 years ago

Same behaviour here with ZHA/Conbee2. Terncy paired and can read battery level, but no events firing. Would be happy to help debug if anyone can give me some directions :)

github-actions[bot] commented 2 years ago

There hasn't been any activity on this issue recently. Due to the high number of incoming GitHub notifications, we have to clean some of the old issues, as many of them have already been resolved with the latest updates. Please make sure to update to the latest version and check if that solves the issue. Let us know if that works for you by adding a comment 👍 This issue has now been marked as stale and will be closed if no further activity occurs. Thank you for your contributions.

labactani commented 2 years ago

i am having the same result with Conbee stick. tried with ZHA >> i can only ready battery level and with Z2M where for i get null value or NA%... Conbee is with the latest firmware as of 13 Jan 2022 - anyone made progress on this?

github-actions[bot] commented 2 years ago

There hasn't been any activity on this issue recently. Due to the high number of incoming GitHub notifications, we have to clean some of the old issues, as many of them have already been resolved with the latest updates. Please make sure to update to the latest version and check if that solves the issue. Let us know if that works for you by adding a comment 👍 This issue has now been marked as stale and will be closed if no further activity occurs. Thank you for your contributions.

martin3000 commented 1 year ago

My terncy dial was working until Dec 2022. Then it stopped sending events. I was using EZSP in december 2022, in january 2023 I switched to ZNP.

MattWestb commented 1 year ago

@martin3000 Did you repairing the devices then merger to ZNP ? If not all devices is trying reporting to the old EZSP IEEE address and the ZNP is having one other so its not working. You can try re adding some controllers and see if its working or not.

martin3000 commented 1 year ago

@MattWestb hi matt, I removed it and repaired it, reconfigured it, nothing helps. After repairing, I see this in the log:

2023-01-22 10:15:19.595 DEBUG (MainThread) [zigpy.appdb] Error handling '_save_attribute' event with (00:0d:6f:00:16:73:b4:77, 1, 0, 4, 'Xiaoyan') params: FOREIGN KEY constraint failed
2023-01-22 10:15:19.603 DEBUG (MainThread) [zigpy.appdb] Error handling '_save_attribute' event with (00:0d:6f:00:16:73:b4:77, 1, 0, 5, 'TERNCY-SD01') params: FOREIGN KEY constraint failed
MattWestb commented 1 year ago

@martin3000 Its looks like ZHA its having problem adding the device in zigbee.db. Try deleting the device and the restarting HA so the DB is being closed and new opened OK. Then (Z)HA is stable try adding the device and see if its being joined OK and also written in the DB.

If its being OK in the DB it shall staying in ZHA after one more restart of HA and if not its being not OK in the GUI (normally unknown device).

martin3000 commented 1 year ago

@MattWestb Thanks Matt for looking into it. But the device is there, see screenshot. I checked the zigbee.db and the tables look good. image

martin3000 commented 1 year ago

I have the same problem now with a osram lightify 4x wireless button: the button was working for 2 years, than the events came double. I reset the button and paired it again, now there are no events at all.

martin3000 commented 1 year ago

Is it possible to activate zigpy/zha/znp debugging to check where the events flow?

martin3000 commented 1 year ago

I think the problem happens when pairing: in the log I see messages that adding the cluster or device to a group did not work. Could it be useful to compare the v9 tables in zigbee.db against the v11 tables for a specific ieee?

Drizzt321 commented 1 year ago

I have the same problem with this device, got to me in the last month from AliExpress, never had one of these. Using ZHA. I can get it to pair, but no zha_events or anything else afterwards.

Here's the log file of when the device joined.

terncy-sd01 join.log

martin3000 commented 1 year ago

Last year it was working, but then suddenly (with a new HA version) it failed.

Drizzt321 commented 1 year ago

@martin3000 And nobody seems to have any idea how to get it to work? What the change that caused it to break?

I'm happy to do any other logging, test custom quirks, etc. This would be the perfect device for my needs, mute/unmute/volume control for each rooms Snapcast powered speakers.

I have another weird issue as well, same Aqara buttons as ones I have now, but these new ones don't produce the same ZHA events I was expecting. Dunno if it's a updated firmware changes something or what, but it's bloody annoying https://github.com/zigpy/zha-device-handlers/issues/2367

tube0013 commented 1 year ago

I added a SD-01 to my network (it was previously paired when I bought them a year or 2 ago) and it is working as expected. I'm using an ezsp coordinator - what are you all using? I can try another one with a test network on a cc2652 coordinator later tonight.

Drizzt321 commented 1 year ago

I'm using one of your older cc2652 ethernet coordinators. Works great for everything else (except those new Aqara Opple, even though they seem exactly the same as ones I already have that are working fine), just this one no zha_events, nothing except the Identify and Battery attributes.

Not sure this is zigpy related, or HA, but I notice that blakadder has an OTA update file there. Tried to update following the instructions, but getting this error stacktrace. This is after changing the conf and uploading the file to the right dir, via File Editor add-on, and doing a HA restart. 

 Home Assistant 2023.4.6 
Supervisor 2023.04.1 
Operating System 10.1 
Frontend 20230411.1 - latest 
2023-05-02 09:50:13.302 ERROR (MainThread) [homeassistant.core] Error executing service: <ServiceCall zha.set_zigbee_cluster_attribute (c:01GZENXE8P6GJSG3E79NE8NEKD): ieee=00:0d:6f:00:16:dc:fb:2e, endpoint_id=1, cluster_id=25, attribute=2, cluster_type=out, value=49502>
Traceback (most recent call last):
  File "/usr/local/lib/python3.10/site-packages/zigpy/device.py", line 317, in request
    return await req.result
asyncio.exceptions.CancelledError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/core.py", line 1838, in catch_exceptions
    await coro_or_task
  File "/usr/src/homeassistant/homeassistant/core.py", line 1857, in _execute_service
    await cast(Callable[[ServiceCall], Awaitable[None]], handler.job.target)(
  File "/usr/src/homeassistant/homeassistant/helpers/service.py", line 874, in admin_handler
    await result
  File "/usr/src/homeassistant/homeassistant/components/zha/websocket_api.py", line 1275, in set_zigbee_cluster_attributes
    response = await zha_device.write_zigbee_attribute(
  File "/usr/src/homeassistant/homeassistant/components/zha/core/device.py", line 666, in write_zigbee_attribute
    response = await cluster.write_attributes(
  File "/usr/local/lib/python3.10/site-packages/zigpy/zcl/__init__.py", line 569, in write_attributes
    return await self.write_attributes_raw(attrs, manufacturer)
  File "/usr/local/lib/python3.10/site-packages/zigpy/zcl/__init__.py", line 575, in write_attributes_raw
    result = await self._write_attributes(attrs, manufacturer=manufacturer)
  File "/usr/local/lib/python3.10/site-packages/zigpy/zcl/__init__.py", line 326, in request
    return await self._endpoint.request(
  File "/usr/local/lib/python3.10/site-packages/zigpy/endpoint.py", line 237, in request
    return await self.device.request(
  File "/usr/local/lib/python3.10/site-packages/zigpy/device.py", line 316, in request
    async with asyncio_timeout(timeout):
  File "/usr/local/lib/python3.10/site-packages/async_timeout/__init__.py", line 129, in __aexit__
    self._do_exit(exc_type)
  File "/usr/local/lib/python3.10/site-packages/async_timeout/__init__.py", line 212, in _do_exit
    raise asyncio.TimeoutError
asyncio.exceptions.TimeoutError
tube0013 commented 1 year ago

I'm pretty sure that is just due not being able to reach the device, you have to make sure it's awake by pressing the button when you do that.

I am the one who contributed that FW, I sniffed it from a packet capture of the terncy hub doing it. So both of my SD-01's are on the newer fw. I will test the other one on cc2652.

Your aqara opple issue should be a separate issue, but you may need to update the coordinator fw.

Drizzt321 commented 1 year ago

Ah! Not able to reach the device, that makes perfect sense. Just did the button push, got the red light, hit Call Service, button push and got the red light again, and no error in the Core logs yet, so I presume it's working. I'll let it alone for 20-30 minutes and see if there's any output anywhere.

Hmm...how do I tell when it's done? Is there a log entry somewhere? A response to the Call Service?

So I did perform an update recently, was moving stuff around, unplugging stuff, changing it around. Had it plugged into a PoE mUSB to power it from my switch. For some reason when I plugged it back in, it wasn't getting or grabbing an IP, or something weird. So I pulled it off, used a FTDI and ESPHomeFlasher and got it into flash mode and flashed whatever the latest from your github is. Unfortunately, there's no version on the web page. I do see zBSL is set to Off though.

And I do have ticket https://github.com/zigpy/zha-device-handlers/issues/2367 already for it, just mentioning it as another issue I've had recently with new ZigBee devices. Hopefully (not holding my breath), it'll be even less when Thread/Matter becomes the dominant platform. I hope. I'll probably pick up your EFR32 PoE in the future, to get both simultaneously.

martin3000 commented 1 year ago

I'm using a CC2652RB ZNP stick, but before I was using a EZSP stick.

tube0013 commented 1 year ago

email me or open an issue on my repo getting off topic here...

for the sd-01 fw update if you enable debug logging for zigpy you should see the update progressing

Drizzt321 commented 1 year ago

For the firmware update, I'm just going to leave well enough alone, since it seems to be going. Just would be nice to know if there was something that'd confirm it was done, so I can try and remove/re-add to see if that fixes anything with the SD-01.

ilarrain commented 1 year ago

I was the one who contributed the original quirk. I did the original testing with an ezsp coordinator, where it works well. Later, I've switched to a Texas Instruments one, and it doesn't. Probably due to differences in coordinator implementation. I think it has to do with the cluster command parsing done in https://github.com/zigpy/zha-device-handlers/blob/dev/zhaquirks/terncy/__init__.py#L131 I assumed some int types that might be wrong, tried to fix them, but it's beyond my limited python/ZigBee debugging/sniffing skills.

Drizzt321 commented 1 year ago

@ilarrain thanks for chiming in, you're probably ahead of me simply because I've never done any sort of ZigBee/quirks stuff directly, just been very fortunate and grateful at all the other folks who have done all the hard work of getting it all more or less (usually more!) working well with HA/FOSS software.

To that end, looking at the ZHA debug log, I see this when I removed, and re-added it. I think you're right, something with the click_event init/handling causes the quirks to fail and stop loading, and thus the device to not usable. 

2023-05-02 17:14:13.989 ERROR (MainThread) [homeassistant.components.websocket_api.http.connection] [140474429676976] Value for field 'state' is required: click_event(count=0)
Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/components/websocket_api/commands.py", line 199, in handle_call_service
    await hass.services.async_call(
  File "/usr/src/homeassistant/homeassistant/core.py", line 1820, in async_call
    task.result()
  File "/usr/src/homeassistant/homeassistant/core.py", line 1857, in _execute_service
    await cast(Callable[[ServiceCall], Awaitable[None]], handler.job.target)(
  File "/usr/src/homeassistant/homeassistant/helpers/service.py", line 874, in admin_handler
    await result
  File "/usr/src/homeassistant/homeassistant/components/zha/websocket_api.py", line 1328, in issue_zigbee_cluster_command
    await zha_device.issue_cluster_command(
  File "/usr/src/homeassistant/homeassistant/components/zha/core/device.py", line 727, in issue_cluster_command
    response = await getattr(cluster, commands[command].name)(
  File "/usr/local/lib/python3.10/site-packages/zigpy/quirks/__init__.py", line 207, in client_command
    return await self.reply(
  File "/usr/local/lib/python3.10/site-packages/zigpy/zcl/__init__.py", line 344, in reply
    hdr, request = self._create_request(
  File "/usr/local/lib/python3.10/site-packages/zigpy/zcl/__init__.py", line 268, in _create_request
    request.serialize()  # Throw an error before generating a new TSN
  File "/usr/local/lib/python3.10/site-packages/zigpy/types/struct.py", line 231, in serialize
    for field, value in self.assigned_fields(strict=True):
  File "/usr/local/lib/python3.10/site-packages/zigpy/types/struct.py", line 177, in assigned_fields
    raise ValueError(
ValueError: Value for field 'state' is required: click_event(count=0)
Drizzt321 commented 1 year ago

Hm, need to see how I can get the raw bytes from the coordinator.

martin3000 commented 1 year ago

I modified zhaquirks/terncy/--init--.py:

   def handle_cluster_request(
        self,
        hdr: foundation.ZCLHeader,
        args: List[Any],
        *,
        dst_addressing: Optional[
            Union[t.Addressing.Group, t.Addressing.IEEE, t.Addressing.NWK]
        ] = None,
    ):
        _LOGGER.debug(
            "Terncy - handle_cluster_request tsn: [%s] command id: %s - args: [%s]",
            hdr.tsn,
            hdr.command_id,
            args,
        )

But in the log there is nothing. So it seems that handle_cluster_request is not called.

martin3000 commented 1 year ago

Going to the device information page and looking at the clusters, I see that the TerncyRawCluster does not have any attributes. Reconfiguring the device gives the following error:

WARNING (MainThread) [homeassistant.components.zha.core.channels.base] [0x5852:1:0x0001]: 'async_configure' stage failed: 'ZCLAttributeDef' object is not subscriptable

The thing is: I have also other buttons here, for example from IKEA, an these button don't work also. I wonder if there is anybody with an EZSP where the terncy button still works?

jfisbein commented 1 year ago

I tried multiple times to upload the firmware but I think it's not working. should the firmware version on the device page change? I always see Firmware: 0x0000001a.

martin3000 commented 1 year ago

I tried multiple times to upload the firmware but I think it's not working. should the firmware version on the device page change? I always see Firmware: 0x0000001a.

image

martin3000 commented 4 months ago

I di not give up hope. Today I removed the Terncy Dial from ZHA, restartet HA and paired it again. Not working. This is the error from the log:

2024-07-30 15:13:44.244 DEBUG (MainThread) [homeassistant.components.zha.core.cluster_handlers] [0x3D63:1:0x0001]: finished cluster handler configuration
[0x3D63:1:0x0001]: 'async_configure' stage succeeded
[0x3D63:1:0x0003]: 'async_configure' stage succeeded
[0x3D63:1:0x0000]: 'async_configure' stage succeeded
[0x3D63:1:0x0020]: 'async_configure' stage failed: Failed to write attribute checkin_interval=13200: <Status.INVALID_VALUE: 135>

No idea what this checkin_interval is.

MattWestb commented 4 months ago

[0x3D63:1:0x0020] = pull control cluster. Its one Zigbee 3 future that good end devices is using for getting pulling working OK with its parent.
But the device dont like getting / setting 135 (i think its out of range then its have to so with slow pull intervals its must being greater then that). Try reading the other attribute on the pull control cluster and see how its looking. And i think then using EZSP Silabs stack is handing it better then the device requesting it and also if is have one other router as its parent and not the coordinator that. Also posting one updated device signature so can see what its shall having for capacity (very likely false Zigbee 3 end device).

Edit: Z2M is saying its only working on channel 20-24.

martin3000 commented 4 months ago

The checkin_interval is 3600 and cannot be changed. But ok II think this is not a problem.

But the TerncyRawCluster is empty, there are no attributes. But I knows the commands "click_event" and "motion_event". This is strange. I think this should be attributes and not commands.

MattWestb commented 4 months ago

3600 sound good (I think its 1/10 or 1/4 seconds) from my experience sniffing IKEA controllers. I think its failing init and the device is using its default but then its not continuing init other functions like the customer cluster.

Is reconfigure from the device card working ? And is it doing all OK no not stopping for not OK settings ?

martin3000 commented 4 months ago

battery_voltage cannot be read, only battery_percentage_remaining. So battery_voltage is yellow. And reconfiguration of PollCluster also fails. TerncyRawCluster is not reconfigured.