Devices constantly become "unavailable" at Home Assistant for a few minites

[SvenAbels]

Version of the integration

latest (v2.13.0)

Configuration

Exactly 20 tapo devices: 18 of them are P100 and 2 of them are P110

Describe the bug

All devices are connected to Home Assistant without any problems and are listing their current poweer consumption. But from time to time some of them become "unavailbale" at Home Assistant. This happens to multiple devices. After 1-5 minutes, the device is typically available again. However, the devices themselve have a proper web connection and I can still see their current valiues (including changes) in the tapo Android app.

In the logs menu of Home Assistant, I can only see that the device becomes unavailable but without any further details about the reason. The devices are connected to the tapo extension using their IP. The IP does not change and the device is still pingable from the PC where Home Assistant is installed on.

I already modified the refresh times of the devices as I was fearing that they might be too low. So I increased them from 10 to 20 seconds but this did not change anything.

The problem does not happen at all times: It sometimes works for many hours without any problem and then the problem appears again - e.g. 10 times an hour for a few hours... I'm experiencing this problem since >4 weeks now (since I started using tapos).

[Elastiek]

Same problem here. When I trie to add the device again in HASSIO it say's "device already configured. Anyone with a solution ?

[SvenAbels]

For me, it works to reload the device by clicking on the 3 dots in the device overview:

However, this is not a good solution because it only works for a while and I cannot reload several devices a day...

[pan07gaz94]

Same problem here too. 11/11 devices not available.

[pan07gaz94]

For me, it works to reload the device by clicking on the 3 dots in the device overview:

However, this is not a good solution because it only works for a while and I cannot reload several devices a day...

For me it didn't working

[SvenAbels]

Same problem here too. 11/11 devices not available.

This might be a different problem because for me, I never have the situation that all are not available at the same time. A typical situation is that 1 or 2 out of 20 devices are unavailable for 1 - 10 minutes or until I reload the config of the device.

Also, in about 95% of the time, all devices are working fine. This unavailability of single devices only happens every now and then. On some days it does not happen at all and on other days it hapens every few minutes that one of the decives gets unavailable and available again.

[ChrisMarriott38]

I get this happening also. Constantly for some. Some are showing the available/unavailable. And others are showing Overheat issues. Then checking and cancelling it as an issue. I've disabled/hidden the Overheat entities as a work around but the power available ones a bit of a pain. For Automatons to detect things. So i just added a 5 minute buffer for checking against it for now.

But it would be nice to have things stable.

[JCummings49]

I get similar behavior with the P125M. The device just turns off periodically during the day. It does not come back on, it must be manually turned on. Very annoying.

[pan07gaz94]

i reconnect all my plugs into my integration but when i restart my HAOS i lost again all my tapo p100 entities.

error : argument of type 'NoneType' is not iterable

[uncle-jegan]

Are your devices accessible in the Tapo app when they "unavailable" in HA? I experimented with isolation of network and DNS records, and get same problem. But when DNS is 8.8.8.8 and network connected directly to internet (devices online in Tapo app) there is no problem with "unavailable" condition. Devices: tapo L530 v1.1.0 build 230721 rel.224802 tapo P115 v1.3.0 build 230905 rel.152200

[ChrisMarriott38]

My actual devices don't become unavailable at all, lights always green also on the sockets. you can see them all the time in the app and home assistant - it's just in the log file that there are overheat and power issues detected. Unless the reconnects are unnoticed.

@petretiandrea

I mean this is silly because this is JUST today. Look how many flipping entries there are!!!!!

I have noticed it seems to also VARY on which device does this. For example. I have 7 sockets currently linked up. It's only that device so far today reporting this. But the other day i had a similar one on another socket doing this.

[SvenAbels]

Are your devices accessible in the Tapo app when they "unavailable" in HA?

Yes my devices are still accessible via the tapo application but not via HomeAssistant during that time. I also get no power consumption values in home assistant during that time but I do the real time power consumption in die App.

[ChrisMarriott38]

All mine are fine in Tapo, nothing is visible as changing. I don't even see any devices states change in home assistant. The only real clue i think is the log book saying something weird happening. Feels like the integration is causing it.

Is there a Tapo logbook anywhere? I'm wondering if it can be compared.

[ChrisMarriott38]

I have moved a Wifi repeater next to one of the plugs which this happened frequently on to diagnose if its a signal strength issue. Will report back.

---

There are some circumstances that Tapo smart devices do not work stably or lost connection from time to time, this could be due to the network environment or some other reasons, this article provides some helpful tips to fix the situation.

Step2 EnsureTapo smart device is receiving a Good Wi-Fi signal by checking RSSI (Received Signal Strength Indicator)

Tapo APP->tap in the left top> Device Settings->Device info ->tap Wi-Fi icon to see signal strength (the value of RSSI)

Strong: more than -50 dBm

Good to Average: among -70 dBm to-50 dBm

Poor: less than -70 dBm

Note: If RSSI is less than -70dBm, the signal is too weak, device disconnections are likely, relocate the smart device or router to get a stronger Wi-Fi signal.

---

One device P110 now has a Wifi repeater in the socket next to it - Signal strength -8(RSSI) So this will help determine if its Stability issue and not to do with the HA Addon componant

[SvenAbels]

Hi,

for me the signal is not the problem. My repeater is in the same room for 4 of the Tapos but they still got this problem from time to time. Also, as I described earlier, I can ping the devices during the unavialability time and their real time data i also visible in the tapo app.

[ChrisMarriott38]

Hi,

for me the signal is not the problem. My repeater is in the same room for 4 of the Tapos but they still got this problem from time to time. Also, as I described earlier, I can ping the devices during the unavailability time and their real time data i also visible in the tapo app.

Agree, i don't think it will likely be to do with this. I'm just going to make certain with this test. I can see the logs, and multiple devices in locations near different wifis' and its happening on a lot. Can you see your logs? Do you get a crazy amount of unavailable/turned off and such like the image i posted above?

Are your devices on 2.4 or 5ghz frequency? That can also be known to be an issue.

[SvenAbels]

My devices are all on 2.4 Ghz

[whoisthedaddy]

Posted this on another issue but might be worth a try...

Like many of you I've had a similar issue and tried a number of fixes (ensuring DNS was mapped to 8.8.8.8 which I read on this post) - none of these worked until I disabled the 'Try to track device dynamic ip using MAC address' (click on 3 dots or Configure and remove check) / as soon as I disabled this all devices came back online. I have a Ubiquiti Dream Machine and all these are on their own IOT VLAN (and were accessible on the app and via pinging) and will now assign them all static IP addresses to see if that helps... Might be useful but might not.... : )

[SvenAbels]

Thanks for the hint. I checked it and the checkbox is already deactiavted on my devices.

I also notices, that the unavailability happens very infrequently: Sometimes I get then 3-4 days in a row every few hours and then sometimes I get no problems for a week or two..

[ChrisMarriott38]

Sadly, use Google DNS, and i have not used the dynamic track part as i already had fixed IP's for each device.

Yea same to the above, its so random and sporadic, sometimes is super frequent.

I did wonder if it could be to do with the device refresh rate. i think mine are default 30s or so.

[whoisthedaddy]

Ok thanks will keep an eye out and comment further - the only two things I’ve changed today are the DNS to 8.8.8.8 and the checkbox. My refresh rate is also set to 30 I believe…

[ChrisMarriott38]

Anyone tried any older versions? To see if its a more recent issue?

[whoisthedaddy]

Nope, I've not tried going back to an older version - wasn't sure whether I'd have to re-add everything which is far too painful! All the re-pairing, adding everything back in and then updating all the entities!

[ChrisMarriott38]

We've not heard anything in regards to this reply. I've updated to the latest home assistant and actually noticed that the TP-Link official integration now detects Tapo devices. It's still early on i think it said so maybe buys? It doesnt detect all the entities like this Tapo integration though.

I can see 4 entities per device on the TP-Link . Switch, current consumption, todays consumption, total consumption.

On this Tapo one i can see 8 entities per device. Switch, Current power, todays energy, Month energy, (These are extras - month run time, signal level, todays run time. overheat.)

I will try disable Tapo, though i would rather use this one as it's seemed better supported before. And see if the log issue clears.

[ChrisMarriott38]

So i tried disable a device from the TAPO integration. Just one of the plugs that reported the most log book errors. And added it via the new TP-LINK Direct integration but i think i get the same log book errors. It seems strange. Unsure if its related to the integrations now or actually if it's more related to the HA build. However, i looked back at my older test HA version.

Home Assistant 2023.8.2 Frontend 20230802.0 - latest

And i see the same Logbook issues in that also.

[ChrisMarriott38]

Anyone see this or similar in their logs?

[SvenAbels]

I did not have this experience for several weeks now. Thus I intent to close this issue unless it's still unresolved for others...

[devangapatel]

I am seeing the same issue/pattern with the Tapo Mini Smart Wifi Plug TP 15. Wifi strength is not a concern. Also the device works fine in their native Tapo app when it is unavailable in HA. The older (4+year old) TP-Link smart plugs do not have this issue but the newer Tapo TP15 is experiencing this problem. This is definitely a bug with HA integration. Can someone please help fix this?

[jaymunro]

I have this issue. Checking to find out when it started...

[jaymunro]

It started before the log pruning date 10 days ago, so no info there. I have 7 devices and the issue is present on all of them with unavailability occurring several times an hour and lasting 5-40 minutes.

I am seeing a lot of these in the log:

This error originated from a custom integration.

Logger: custom_components.tapo.coordinators
Source: helpers/update_coordinator.py:347
integration: TP-Link Tapo (documentation, issues)
First occurred: 11:34:17 AM (28 occurrences)
Last logged: 12:07:26 PM

Error fetching tapo data: Unexpected exception:

Note HA was restarted at 11:33am

I have enabled debug logging to try and get more detail.

[jaymunro]

Not seeing anything more helpful in the DEBUG entries:

2024-04-23 12:17:35.803 ERROR (MainThread) [custom_components.tapo.coordinators] Error fetching tapo data: Unexpected exception: 
2024-04-23 12:17:35.803 DEBUG (MainThread) [custom_components.tapo.coordinators] Finished fetching tapo data in 10.003 seconds (success: False)

[elliotcz97]

Same Issue still happens, for me it's a bit better because it reconnects very fast, but still annoying.

[LaszloV369]

Hello everyone. I have found a solution, at least for me it works. I have 10 tapo devices one or two became unvailable randomly, very annoying!!! I have observed that the wifi channel on the router was set to Auto. I mooved it to manual setting and set to be always on channel 6. I chose this cahannel using wifi analyzer android app on the phone to see wich channel is the most "empty" - no neighbouring wifis to interfere. Since then I hand NO MORE unavailability issues on any of the devices! My router is a Tplink Archer x73. How this helps!

[hareenl]

I'm also seeing this issue and it happens constantly through out the day accross 11 Tapo P110 outlets. It sometimes happens to groups of outlets and only certain outlets at other times. I have strong WiFi across the house with 3 access points on channels 1,6 and 11. All devices on the network are pinged utilizing uptime kuma and nothing drops on the network itself. The following was seen on the HA logs.

[jaymunro]

It used to happen to me a lot. Then I completely uninstalled the HACS version - deleted all entities and deleted the HACS integration version. That was months ago and it's only happened a couple of times since, usually after a significant upgrade in OS or HA. Suggest you try that or say that your only running on the HA Core integration and have removed the HACS one (which doesn't seem to be getting any love).

[andyevans-eng]

Since adding a Tapo P110 to my network one of my Kasa KP115 plugs occasionally becomes unavailable, but it was rock solid before, and the P110 becomes unavailable every 5 minutes or so. Signal strength is excellent. Only started having problems when P110 was set up. While devices are unavailable, they are still online in the Tapo app. It is getting very frustrating as my energy data in HA is intermittent because of it being unavailable, but all data in the Tapo app

[zqnikals]

same issue. I have 11 plugs for a few weeks. the signal is strong. today they became to become unavailable one by one. Also, i found One strange thing: sometimes plugs set auto-off timer to 120. I have never had such automation.

[Kopetan4egX]

sometimes plugs set auto-off timer to 120. I have never had such automation.

Auto-off timer time (120 by default for me, and seems like for you too) and the timer (meaning on/off status) itself are separate entities. The one with seconds just shows the time before turning of, it won't turn off the socket, if the entity "Auto-off enabled" is off/

[s55ma]

Also the same problem, but with Tapo P115 plugs. Kasa plugs works fine and other non Tplink devices also work fine.

[c-kick]

This is still an issue as of version 2024.8.3. And still, the only workaround to get things working again is blocking internet access to all the P115 plugs. But that doesn't even restore everything, as some of the device entities will still not work (such as current and total power usage, which is one of the main reasons I invested in these plugs).