deconz-rest-plugin
deconz-rest-plugin copied to clipboard
dresden-elektronik
Lumi.plug.mmeu01 no more power consumption
Describe the bug
Good morning Since I updated to version 2.18.02 / September 19, 2022, no more power consumption is transmitted from my sockets. There were no problems with version v2.17.1. Can someone check this please.
Steps to reproduce the behavior
updated to version 2.18.02
Expected behavior
Screenshots
Environment
- Host system: Raspberry Pi
- Running method: Raspbian
- Firmware version: 26780700
- deCONZ version: 2.18.02
- Device: RaspBee II
deCONZ Logs
Additional context
HerstellerLUMI ModellZwischenstecker Version 09-06-2019
Hi,
It's impossible for us to understand what sockets you mean. Either way, i see something with "lumi", assuming this mightbe related: https://github.com/dresden-elektronik/deconz-rest-plugin/issues/6340
Hi; It's about these values that are no longer updated, it seems as if these data points are virtually non-existent. I use ioBroker with the Deconz adapter to read the data, which has always worked so far.
https://www.amazon.de/Xiaomi-Ferngesteuerte-elektronschen-Sprachassistent-Stromverbrauchsmessung/dp/B083F91ZX3
Hi,
CAn you check the basic cluster in deconz? What model ID / Manufacturere name are these? https://github.com/dresden-elektronik/deconz-rest-plugin/wiki/How-to-read-Clusters
so I have another plug with the same model identifier lumi.plug.mmeu01 because data always arrives. Have now deleted the plug and retrained now a lot of data points are created but it takes time to change. or update.
so ich habe noch einen Plug mit dem selben Model Identifier lumi.plug.mmeu01 da kommen Daten immer an. Habe jetzt einmal den Plug gelöscht und neu angelernt jetzt werden jede Menge Datenpunkte angelegt aber es dauert bis sich sich ändern. oder aktualisieren.
Hi checked again, but the rest of my ZigBee sockets no longer provide any data on current consumption or amps.
So noch mal geprüft, der Rest meiner ZigBee Steckdosen liefern aber keine Daten mehr über aktuellen Verbrauch oder Ampere
Hi there,
I have the same problem, I did some investigations, I wanted to post also, but I found this thread, so I will provide here some info.
I have 5 devices, of type lumi.plug.mneu01. Out of 5, one is still providing info (sensors exist in API), the rest no.
While looking in the docker logs, I found the following relevant lines:
02:40:27:728 DB skip loading sensor Power 24 24, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:728 DB skip loading sensor Consumption 25 25, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:729 DB skip loading sensor Power 2 2, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:729 DB skip loading sensor Consumption 3 3, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:729 DB skip loading sensor Power 34 34, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:729 DB skip loading sensor Consumption 35 35, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:729 DB skip loading sensor Power 22 22, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:730 DB skip loading sensor Consumption 23 23, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:730 DB skip loading sensor Power 31 31, handled by DDF Mijia smart plug ZNCZ04LM
02:40:27:730 DB skip loading sensor Consumption 32 32, handled by DDF Mijia smart plug ZNCZ04LM
A bit down, we have the following lines
02:40:27:904 skip legacy loading 04:cf:8c:df:3c:7e:3f:20-01 / lumi.plug.mmeu01
02:40:27:905 skip legacy loading 04:cf:8c:df:3c:7e:3f:20-01 / lumi.plug.mmeu01
02:40:27:905 skip legacy loading 04:cf:8c:df:3c:7e:3f:20-01 / lumi.plug.mmeu01
02:40:27:906 skip legacy loading 04:cf:8c:df:3c:7e:3f:20-01 / lumi.plug.mmeu01
02:40:27:906 skip legacy loading 04:cf:8c:df:3c:7e:3f:20-01 / lumi.plug.mmeu01
It is the id (mac?) of the only device that is still providing info and it is 5 times. The rest of ids do not appear in this format.
Then, this one, an id for a non reporting device:
02:40:28:090 DEV no DDF for 0x04CF8CDF3C7E52B3, modelId: lumi.plug.mmeu01
02:40:28:090 DEV create on-the-fly DDF for 0x04CF8CDF3C7E52B3
or another one, an id for a non reporting device:
02:40:28:122 DEV found DDF for 0x04CF8CDF3C7E41E6, path:
And even down, the device is 0x04CF8CDF3C7E52B3 pooled:
02:40:50:267 0x04CF8CDF3C7E52B3 extract Xiaomi special attribute 0x00F7
02:40:50:267 64 on/off 1
02:40:50:268 03 Device temperature 31 °C
02:40:50:268 98 power 1.663000 (2)
02:40:50:268 95 consumption 124.085373 (124085)
02:40:50:269 96 voltage 2300.000000 (230)
02:40:50:269 97 current 7.230434 (7)
02:40:50:269 05 RSSI dB (?) 8 (0x0008)
02:40:50:270 9a unknown 0 (0x00)
02:40:50:270 08 unknown 278 (0x0116)
02:40:50:270 07 unknown 0 (0x0000000000000000)
02:40:50:270 09 unknown 3073 (0x0C01)
02:40:50:271 0b unknown 0 (0x00)
02:40:50:271 9b Consumer connected (yes/no) 1
02:40:50:271 ZCL attribute report 0x04CF8CDF3C7E52B3 for cluster: 0xFCC0, ep: 0x01, frame control: 0x1C, mfcode: 0x115F
02:40:52:975 0x04CF8CDF3C7E5423 extract Xiaomi special attribute 0x00F7
02:40:52:975 64 on/off 1
02:40:52:976 03 Device temperature 30 °C
02:40:52:976 98 power 1.716000 (2)
02:40:52:976 95 consumption 230.796967 (230797)
02:40:52:977 96 voltage 2230.000000 (223)
02:40:52:977 97 current 7.695067 (8)
02:40:52:977 05 RSSI dB (?) 8 (0x0008)
02:40:52:978 9a unknown 0 (0x00)
02:40:52:978 08 unknown 278 (0x0116)
02:40:52:978 07 unknown 0 (0x0000000000000000)
02:40:52:979 09 unknown 4358 (0x1106)
02:40:52:979 0b unknown 0 (0x00)
02:40:52:979 9b Consumer connected (yes/no) 1
and
03:10:32:296 0x04cf8cdf3c7e52b3 found group 0xFFF0
and
03:12:21:488 ZCL attribute report 0x04CF8CDF3C7E52B3 for cluster: 0xFCC0, ep: 0x01, frame control: 0x1C, mfcode: 0x115F
03:12:25:506 binding for cluster 0x0006 of 0x04CF8CDF3C7E52B3 exists (verified by reporting)
03:12:25:508 skip configure report for cluster: 0x0006 attr: 0x0000 of node 0x04CF8CDF3C7E52B3 (seems to be active)
For the other one,
02:41:48:080 0x04CF8CDF3C7E41E6 extract Xiaomi special attribute 0x00F7
02:41:48:081 64 on/off 1
02:41:48:081 03 Device temperature 28 °C
02:41:48:081 98 power 15.906000 (16)
02:41:48:081 95 consumption 811.153931 (811154)
02:41:48:081 96 voltage 2250.000000 (225)
02:41:48:082 97 current 70.693336 (71)
02:41:48:082 05 RSSI dB (?) 8 (0x0008)
02:41:48:082 9a unknown 0 (0x00)
02:41:48:082 08 unknown 278 (0x0116)
02:41:48:082 07 unknown 0 (0x0000000000000000)
02:41:48:082 09 unknown 2816 (0x0B00)
02:41:48:082 0b unknown 0 (0x00)
02:41:48:083 9b Consumer connected (yes/no) 1
02:41:48:083 ZCL attribute report 0x04CF8CDF3C7E41E6 for cluster: 0xFCC0, ep: 0x01, frame control: 0x1C, mfcode: 0x115F
02:41:49:507 binding for cluster 0x0006 of 0x04CF8CDF3C7E41E6 exists (verified by reporting)
02:41:49:509 skip configure report for cluster: 0x0006 attr: 0x0000 of node 0x04CF8CDF3C7E41E6 (seems to be active)
If you need full logs, or any other info, I am ready to provide them.
Thank you for your time.
Maybe it helps (although I think it is because of ddf), in VNC they look different - 0x04CF8CDF3C7E3F20 is the one that reports:
Hello, I can confirm that's the case with me too, with the exception of one, the rest of the sockets no longer send anything via deconz.
Please refer to https://github.com/dresden-elektronik/deconz-rest-plugin/wiki/Xiaomi-smart-plugs-not-working-correctly
Well after updating I had some decimal issues in the home assistant addon. Other reported it here
https://github.com/home-assistant/addons/issues/2679
I updated the firmware according to the link. It seems to fix the decimal issue form the other topic but added another. It seems that the power state is not temp. saved:
Our tumble dryer uses approx 600w fully laoded so the "peaks" seams to be plausible. But then immeadiately drops to 0w.
Also missing a consumption entity in Home assistant. Not sure if this is a error from the rest api or the HA Addon/Integration
Dis voltage seems to correlate with the peaks
15:35:22:824 0x54EF441000431249 extract Xiaomi special attribute 0x00F7 15:35:22:825 64 on/off 1 15:35:22:825 03 Device temperature 27 °C 15:35:22:825 98 power 253.050003 (253) 15:35:22:826 95 consumption 0.571267 (571) 15:35:22:826 96 voltage 2280.000000 (228) 15:35:22:827 97 current 1109.868408 (1110) 15:35:22:828 05 RSSI dB (?) 2 (0x0002) 15:35:22:828 9a unknown 0 (0x00) 15:35:22:828 08 unknown 297 (0x0129) 15:35:22:829 09 unknown 2049 (0x0801) 15:35:22:829 0b unknown 0 (0x00) 15:35:22:830 9b Consumer connected (yes/no) 1 15:35:22:830 0a Parent NWK 0 (0x0000) 15:35:22:831 0c unknown 1 (0x01) 15:35:22:831 ZCL attribute report 0x54EF441000431249 for cluster: 0xFCC0, ep: 0x01, frame control: 0x1C, mfcode: 0x115F 15:35:23:315 Device TTL 6419 s flags: 0x7 15:35:26:367 ZCL attribute report 0x54EF441000431249 for cluster: 0x000C, ep: 0x15, frame control: 0x1C, mfcode: 0x115F 15:35:36:551 ZCL attribute report 0x54EF441000431249 for cluster: 0x000C, ep: 0x15, frame control: 0x1C, mfcode: 0x115F
Hello, I'm getting too old for this. Dealing with it at 60 is complicated. No firmware is displayed for me, no matter which device I take. Resetting did not bring any real success. It takes forever until something changes, for example with the current consumption. Another problem is that new data points are now being created that do not have the original name, see the picture. So if I take the name of the washing machine, sensor values are shown to me with other names. Now how am I supposed to find out what's what?
Grüße Fabio danke für die Hilfe
@SwoopX is this a deconz implementation issue or a device bug / user issue. If it's the latter, I'd rather close this and forward users to the forums.
Well it was introduced when updating deCONZ, and is fixed by rolling deCONZ back again. So it is a deCONZ implementation issue.
Since my Problem only maybe relates to the same topic here and I am not sure if the Problem is from "deCONZ", "deCONZ-rest-plugin" or the "home assistant Integration" I created a forum post: https://forum.phoscon.de/t/aqara-powersocket-consumption-issues/2494
@SwoopX is this a deconz implementation issue or a device bug / user issue. If it's the latter, I'd rather close this and forward users to the forums.
Well, as it can be read in the original issue, the device seemed to be jumpy right from the beginning on what it does expose. No fixed pattern to reproduce that and quite some attempts to press it into one scheme to have it somehow work. I always suspected the firmware to have issues, though I never mentioned in the issue.
With the new device firmware available since a few months, all that jumping around has stopped as far as I can tell based on my tests and it is finally exposed in one fixed state.
Any plug now making trouble was never meant to be allowed to play with deconz in the first place and apparently the measures to block that were not as successful as they should have been.
Hi all!
I can confirm that ALL Lumi Plugs (lumi.plug.mmeu01) dont provide Power and Consumption any more. The ONLY thing that changes is updating Homeassistant addon deconz to version 6.16.0 (updating vom 6.15.0). From this time this addon was updated i only get the power state visible in homeassistant (on(off).
Product: Conbee II Version: 2.18.02 / 9/19/2022 deconz Firmware not changed with this update: version 26720700
Running on: Home Assistant 2022.9.6 Supervisor 2022.09.1 Operating System 9.0 Frontend 20220907.2 - latest
Buat de deconz log is providing this: 11:46:37:682 0x04CF8CDF3C8A37D6 extract Xiaomi special attribute 0x00F7 11:46:37:682 64 on/off 1 11:46:37:682 03 Device temperature 27 °C 11:46:37:683 98 power 8.819000 (9) 11:46:37:683 95 consumption 31.422968 (31423) 11:46:37:683 96 voltage 2310.000000 (231) 11:46:37:683 97 current 38.177490 (38) 11:46:37:684 05 RSSI dB (?) 2 (0x0002) 11:46:37:684 9a unknown 0 (0x00) 11:46:37:684 08 unknown 278 (0x0116) 11:46:37:684 07 unknown 0 (0x0000000000000000) 11:46:37:685 09 unknown 2816 (0x0B00) 11:46:37:685 0b unknown 0 (0x00) 11:46:37:685 9b Consumer connected (yes/no) 1
I dont understand why this values are not shown in homeassistant:
I think its perhaps a change of source code in the addon thats not being considered for the integration.
config_entry-deconz-a198b989cee311b87b9589379e8b448b.json.txt
config_entry-deconz-a19448b.json.txt
A revert from backup to version 2.17.1 / 7/1/2022, HA addon 6.15.0 bring all sesnors back:
Conclusion: Its a 100% bug of the last deconz addon version 6.16.0.
Also have the same problem. Updated deconz yesterday through Home Assistant and since then I don't get any meter sensors anymore. Only on/off works so definitely a bug that is caused by the deconz update.
How can I go back to the previous version btw?
Also have the same problem. Updated deconz yesterday through Home Assistant and since then I don't get any meter sensors anymore. Only on/off works so definitely a bug that is caused by the deconz update.
How can I go back to the previous version btw?
If you made a backup previously as recommended by the update process (defult activated) go to /settings/system/backup
Go to restore and then restart Homeassistant. Thats all
Also have the same problem. Updated deconz yesterday through Home Assistant and since then I don't get any meter sensors anymore. Only on/off works so definitely a bug that is caused by the deconz update. How can I go back to the previous version btw?
If you made a backup previously as recommended by the update process (defult activated) go to /settings/system/backup
Go to restore and then restart Homeassistant. Thats all
Thanks, I was looking at Deconz website but should have looked here lol. Anyway I'm back on 2.17.01 / 7/1/2022 in Home Assistant and the meters are working again. Thanks!!
Same here, rolling back solves the issue. I guess we'll have to wait for a new update before doing anything else.
So, I did the FW update on only one device (just to test) and weird things started to happen. So, from my memory, it was something like:
- after the FW update, in VNC the device looked ok, but no sensors (re)appeared (consumption and power)
- after a deconz restart, the device disappeared from VNC, but one like this appeared (it is not the one, just example) - note the missing second dot:
- ok, I said, let's try another inclusion, so, add new lights, reset the device, it reappeared correctly (includin the name, it changed back to 'Bedroom plug etc', and only one of the sensors (I think the total consumption was missing)
- another removal, readd, deconz restart, now it lloks ok in VNC, but I do not have any sensors
Buuut... Somewhere on the way, another device started to report the Consumption, Power (sensors), although I did not upgrade the FW on it.
So, I give up and I ordered a CC2652P and I will switch to ZHA in HA (at least for the plugs in the beginning).
I saved the deconz docker logs; if they might help you, just let me know.
Cheers
Same issue with Xiaomi smart plugs. I have rolled back to 6.15.0 and the problem disappeared.
@raoulx24 can you please confirm after you added the plug with the new FW works in ZHA? I am considering same thing...
I have the same issue. Updated to 2.18.02 and my power meter shows 0 w. Rollback to 2.17.01 solve my issue.
Copy pasta from related: https://github.com/dresden-elektronik/deconz-rest-plugin/issues/6372
The wiki has a few remarks for the device in https://github.com/dresden-elektronik/deconz-rest-plugin/wiki/Xiaomi-smart-plugs-not-working-correctly from what I understand the plug should work with the DDF when it's updated to the recent OTA firmware. The older Xiaomi firmware is a bit messy.
Side note: The following PR adds the "known issues" section with a link to this wiki in the DDF in https://github.com/dresden-elektronik/deconz-rest-plugin/pull/6386 to show up in http://127.0.0.1:8000/devices/xiaomi/xiaomi_zncz04lm_smart_plug_v24/
Yeah as mentioned I did this FW Update and totally regret it. Since then I have only more issues with my aqara smart plug. Luckily It's Not only me, so hoping for a fix. Parallel ordered a sonoff zigbee stick from aliexpress either (soft) switching to ZHA or downgrading smart plug firmware. Before Firmware Update I only had 10x readings which just could be adjusted manually. https://forum.phoscon.de/t/aqara-powersocket-consumption-issues/2494/5
@Mask3007, after a couple of hours in zigbee2mqtt (I decided to go the hardcore way), two plugs, one older FW and one latest look ok, so no need to 'downgrade' the FW is needed (and, btw, I do not think it is possible).
I will not start here to make a comparison of deconz and zigbee2mqtt, as I consider that both products are solid and I respect a lot the work done in deconz/phoscon/conbee (I am a devops manager in a cloud company, so I consider that I can have a pertinent opinion)... But, in this specific case, I think it is not handled properly here.
I also had some Aqara Power Plugs and I remember that they started to behave strangely in Consumption reporting, so I changed them with other ones.
As I general and final comment, I can understand the problems in zigbee and especially in deconz. The approval process seems to be way too permissive, there are so many manufacturers which implement the protocol more or less to the letter (some of them seems like they think that they invented the wheel or implement some strang things to be sure that you will use only their ecosystem - here I am looking to Tuya, mainly... At least compared with zwave (which I also use in my house), where all this problems really do not exist. Zigebee vs ZWave is like comparing Android world to Apple (and I use a Google Pixel device, so some might say that I am somewhere in the middle)
Cheers and good luck to all of you
I also upgraded my plugs and I also regretted it at first, but I managed to figure out the method to reset and re-pair the device and I'm happy with the upgrade. For me the problem is not with the deConz but with the plugs that require some patience to re-pair. I used the method for about 15 plugs and everything is working perfectly with DDF version 24 for these plugs. If anyone is still interested I can leave here the steps I did, it might help someone.
