Tibber Pulse data (polling) seems not recover by itself after Tibber Pulse is back on-line after been off-line for a while.

[RicardP]

The problem

Power consumption etc available by Tibber Pulse (P1) Wifi dongle, via Tibber cloud service to HA, wont recover without MANUALLY RELOADING Tibber integration if the Pulse dongle been off-line for sertain time.

What version of Home Assistant Core has the issue?

2022.8.2

What was the last working version of Home Assistant Core?

No response

What type of installation are you running?

Home Assistant Container

Integration causing the issue

Tibber

Link to integration documentation on our website

No response

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

No response

Additional information

No response

[alexanderwink]

This is also true if the Tibber API does not respond for some reason (down) or in case of a 429 response code.

[probot-home-assistant[bot]]

Hey there @danielhiversen, mind taking a look at this issue as it has been labeled with an integration (tibber) you are listed as a code owner for? Thanks! _{^{(message by CodeOwnersMention)}}

---

tibber documentation tibber source _{^{(message by IssueLinks)}}

[knutfh]

It seems to also be true if Home Assistant has an unreliable Internet connection. I'll try to take a look at it (it happens daily at my place).

[ZyberSE]

This is happening frequently for me too, would be nice if there was a way to resolve it.

[Ontrackx]

Happening with me to, have to reload intrigation all the time

[StoricU]

Started to happen to me too, began for about a week ago. Have to manually reload the integration for it to work again.

[large]

Same problem here, no unstable connection in my end. Don't have any logging to help with atm

[Wilsson2000]

Same here, have to reload almost every day for a couple of weeks now.

[imist]

Same for me, happened 2022-10-20 21:48:58 and 2022-10-22 23:05:31 Central European Time. No sign of error in ordinary log. Manual reload solves problem.

[erikosvaldsson]

Same for me, fails roughly every other day and requires a restart of home assistant.

[imist]

FWIW: happened again last night at 2022-10-23 22:07:48 CET.

[mflage]

Don't know how useful this is, but the same thing happened for me at around exactly the same time - 22:07:24 CET. Then I got a a few readings from 22:20:04 to 22:22:24 and then it was "permanently" down and I had to reload the integration and then it went down again at around 02:02:25. So the theory that it doesn't handle a temporarily outage from Tibber seems reasonable.

[knutfh]

I too had an outage at 2022-10-23 22:07:24 CET. This is likely due to Tibber not responding for some reason, as suggested my @mflage. Blueprint automations that reload the service does not seem to have any effect - I have to manually reload the integration to get it running again.

[mflage]

Blueprint automations that reload the service does not seem to have any effect - I have to manually reload the integration to get it running again.

Just had an outage a few minutes ago (10:05:00), but my automation was able to recover from it:

alias: Reload Tibber if we have stale readings
description: ""
trigger:
  - platform: template
    value_template: >-
      {{ now() - states.sensor.<entity>.last_updated >
      timedelta(minutes=5) }}
action:
  - service: homeassistant.reload_config_entry
    data: {}
    target:
      entity_id: <entity>
mode: single

This automation will reload the integration referenced by entity_id if .last_updated hasn't updated in 5 minutes. I also want to expand on this automation to use all four Tibber devices as input - as it's of course valid that one of them loses internet connectivity for any number of reasons, and that shouldn't be enough to just reload the whole integration.

[VidarJuvik]

Same problem as listed above. I've restarted integration several times today, and after the last reset it worked for only a few minutes. My GF has the same setup, but strangely it works just fine for her.

[whohencode]

Also seen this issue. Tibber sensors stalls, CPU usage goes high, restart integration and it works again.

[torge70]

Tibber integration is very unstable lately. Stops every day.

[strelniece]

Same problem here and restart the integration if no new report in 5 minutes (Flow in Node-RED).

Is there any additional data that we can provide to help narrow down the problem?

[enricojonas]

Hi, same issue for me since a couple of days back. CPU goes 100 % and I need to restart the integration manually, no data received.

Is anyone looking into that? Unfortunately not much in the logs, only thing I can see is:

2022-10-26 06:26:34.583 ERROR (MainThread) [homeassistant] Error doing job: Task was destroyed but it is pending!

Around the time when this happens, don't know if it is connected though.

[tfsjohan]

Have the same problem. One with Watty and one with Pulse. Both have stable internet connections. Hope this gets fixed soon as most of my use of HA is dependent on the Tibber integration.

[RicardP]

@Danielhiversen Did you note this issue? I am sorry but I think we need your help here to find a solution, many users have same critical problem.

Thank you for your support and nice work!

[Danielhiversen]

It is on my todo list, but currently not a high priority for me to spend time on this

[enricojonas]

It eats kWh as CPU goes 100% ...

[mflage]

I think we need to try and figure out why this is not a more common problem than 23 comments reflect. At first I thought I had this problem due to some rate-limiting, as I was polling the Tibber API from two different HA installations. But after disabling the second HA and I still was experiencing this issue, I went searching for answers and stumbled across this - rather old - thread.

I suspect this doesn't happen to all of the users of the Tibber integration, so maybe we should try and find the common factor for all of us? I'm myself using Tibber on three different locations where I have a subscription to their electricity as well and then a fourth and last location where I've only added a Tibber Pulse, but where I'm only using this for monitoring. I have added the last location after these issues started, so it's not related to that.

I don't have the exact date and time for when I first experienced this, but I suspect it's been going on for at least 3 or 4 weeks, maybe even longer.

[kklem0]

I think we need to try and figure out why this is not a more common problem than 23 comments reflect. At first I thought I had this problem due to some rate-limiting, as I was polling the Tibber API from two different HA installations. But after disabling the second HA and I still was experiencing this issue, I went searching for answers and stumbled across this - rather old - thread.

I suspect this doesn't happen to all of the users of the Tibber integration, so maybe we should try and find the common factor for all of us? I'm myself using Tibber on three different locations where I have a subscription to their electricity as well and then a fourth and last location where I've only added a Tibber Pulse, but where I'm only using this for monitoring. I have added the last location after these issues started, so it's not related to that.

I don't have the exact date and time for when I first experienced this, but I suspect it's been going on for at least 3 or 4 weeks, maybe even longer.

I spotted this error but it might be fixed by upcoming 2022.10.6. https://github.com/Danielhiversen/pyTibber/issues/207 But I haven't tested it yet.

[anhaff]

Happens to me a few times a day but repaira often on its own. I realize it most for the "consumption current hour" sensor. Today I made the attached screen dump:

(In the end it is the manual restart)

[enricojonas]

Just happened again to me:

Logger: homeassistant
Source: runner.py:119
First occurred: 1:07:28 AM (30 occurrences)
Last logged: 3:22:57 PM

Error doing job: Task was destroyed but it is pending!

I have enabled more logging now, let's see next time it happens.

[NiklasRamstrom]

Danielhiversen/pyTibber#207

My guess is that this is not due to rate limits. This data (power and accumulated power etc) are received via a graphql subscription, hence pushed from Tibber. I would guess that rate limits only applies to polling data via graphql queries.

My guess is that the subscription is lost, for some reason, and the home assistant implementation does not handle this.

And, by the way, I also need to restart my HA tibber integration almost every day, since the "current power consumption" does not update.

[sportopeden]

Same for me. Until it's fixed I'm trying an automation to reload Tibber if processor use exceeds a certain percent.

[ZyberSE]

My cpu % does not seam to be affected, it stays the same (around 20%) even when Tibber Pulse stops working.

[mumse]

I have this issue as well. Sometimes polling works for days, sometimes just a few hours after reloading the integration (or restarting HA).

[enricojonas]

I got some more logs now, maybe this helps?! After this, CPU goes 100%.

Logger: tibber
Source: /usr/local/lib/python3.10/site-packages/tibber/__init__.py:135
First occurred: October 26, 2022 at 11:36:30 PM (1 occurrences)
Last logged: October 26, 2022 at 11:36:30 PM

Timed out when connecting to Tibber

Logger: homeassistant.components.tibber.sensor
Source: helpers/update_coordinator.py:151
Integration: Tibber (documentation, issues)
First occurred: October 26, 2022 at 11:36:30 PM (1 occurrences)
Last logged: October 26, 2022 at 11:36:30 PM

Timeout fetching Tibber XXX data

[MilletGtR]

Happened to me as well. Sometimes twice per day. Other days it works flawlessly.

[steffo007]

I have the same problem as most others. But since yesterday it has gotten worse, now it only works for short moments, from minutes to hours, then I have to reload the configuration for it to start again. 2022-10-26 stopped at 23:38 +1 (CET + 1 hour (standard time). CEST = UTC + 2 hours (summer time))

[StoricU]

The last 24 hours it has not worked.

[anhaff]

Strange.... As of writing this, I never saw this message before.

And also the current power value sensor freezes. This does not look as if Tibber made an (intentional) reduction of service.

[Lunkobelix]

I have the same issue. Using a Tibber Pulse P1 but I don't buy the electricity from Tibber. Started 1-1,5 month ago and happened about once per week, but the last 24 hours it has happened 4 times. Reloading the Tibber integration solves it until next issue occurs. Is it Pulse that is the issue here?

[eirikgrindevoll]

Same error here. I think we can say its an error for everyone?

[anhaff]

Sometimes (when this first happend to me) I checked with the Tibber app and that worked. This should mean that it is the integration, unless the API serves the app different from API-use by the integration

[kilrolf]

This issue has been going on for a couple of weeks now (after the last update of the core?), I decided to use node-red and the node: node-red-contrib-tibber-api and feed everything into my MQTT server. Perhaps not a solution for everyone, but at least it has been stable. I too suspect somehting is going on with the integration, seeing that the Node-red node is very stable.

[Lunkobelix]

Is there no way of reloading the Tibber integration by calling a service? My CPU always freezes at 25% when this happens so an automation would be a simple (and temporary) solution if I just knew how to reload the Tibber integration.

[sportopeden]

This is what I do, I think it works ok:

alias: Reload Tibber description: "" trigger:

• platform: numeric_state entity_id: sensor.processor_use above: 25 action:
• service: homeassistant.reload_config_entry data: {} target: entity_id:
  • sensor.power_xxxxxx mode: single

[Lunkobelix]

Alright, thanks! So reloading one entity is enough? Which one is it that you are chosing? The sensor.power_?

[kklem0]

This is what I do, I think it works ok:

alias: Reload Tibber description: "" trigger:

• platform: numeric_state entity_id: sensor.processor_use above: 25 action:
• service: homeassistant.reload_config_entry data: {} target: entity_id:
  • sensor.power_xxxxxx mode: single

I'm not so sure if there only reload Tibber

[sportopeden]

Yes. Still evaluating, but it has triggered a few times and seems to be doing ok.

[ismarslomic]

I have had issues with 100% CPU usage since 10.10.2022 with my HA running in Docker. Suddenly CPU load goes crazy and the only way to reduce it is to restart the container. After many days with manual damage control I inspected CPU processes today and identified that it is caused by the graphql_subscription_manager. You can find the CPU diagnostic details in this HA Forum thread.

This is quite severe issue and impacts complete HA installation! I hope you can fix it soon. Im willing to help you out testing the bugfix when its ready!

[Bokenbergen]

@Danielhiversen I've had this issue for a while now as well. Think you need to prioritize this issue since more and more people are getting frustrated with it. Creating an automation to reload the integration is not a fix.

[StoricU]

I have the same issue. Using a Tibber Pulse P1 but I don't buy the electricity from Tibber. Started 1-1,5 month ago and happened about once per week, but the last 24 hours it has happened 4 times. Reloading the Tibber integration solves it until next issue occurs. Is it Pulse that is the issue here?

I also use tibber on Homeseer, and there are no issues on that end, so it is not a Pulse problem.

[ismarslomic]

From my investigation on the high CPU load I can see that there is some issues with GraphQL calls in pyTibber lib. I got also help from @bdraco and he gave following input:

It looks like you have something that is reading from a websocket that is either closed and it doesn’t know and keep reading over and over or the websocket is flooding it with data

I have also recognised that some Tibber entities where unavailable or did not get updated, but thats not critical for my automations. What has caused me a lot of trouble is the increased CPU load, which makes the whole HA instance unstable, in addition to increase the heat of the CPU.

[Danielhiversen]

@ofdplz

Think you need to prioritize this issue since more and more people are getting frustrated with it.

Why do I have to prioritize this? Such a comment makes me less motivated to prioritize it.