Error inside async loop:

16-10-22 21:58:30 homeassistant.core: Error inside async loop: Task exception was never retrieved

After updating to 0.31.0 I am getting this error repeatedly in my log. I have never seen this before.
Any ideas?

I am in the same boat. I will be spending some time trying to track down the source of this. I am sure it’s just something in one of my config files…finding which one will be a challenge.

I noticed something that may help track down the culprit: the errors come at regular, repeated intervals. They fire at 01, 10, 20, 32, 40, 50 seconds and that holds steady for over two hours of log entries. Running Hassbian image 1.0 updated to 0.31.0 (in the log below I shortened homeassistant.core to ha.core so it would fit on a single line)

16-10-23 09:10:01 ha.core: Error inside async loop: Task exception was never retrieved
16-10-23 09:10:10 ha.core: Error inside async loop: Task exception was never retrieved
16-10-23 09:13:50 ha.core: Error inside async loop: Task exception was never retrieved
16-10-23 09:14:01 ha.core: Error inside async loop: Task exception was never retrieved

If I discover any useful information I’ll post it here and if it looks like a bug (instead of a wonky config) I’ll raise an issue on GitHub.

I will dig in too. Are you using the scan_interval: parameter on any of your entities? I added that to one of mine just before the update.

^^^Removing that did not help^^^

Don’t think so but I’ll check.

I have been struggling with this all day. I found some sloppy yaml in a couple of my configs (maybe this update is more strict with yaml?). Although tidying the YAML allowed me to control my devices again (yay!), I am still unable to get my scripts to load. I even manually merged my my script files, validated the whole thing with lint and they still refuse to load. Without the scripts only a handful of my automation works.

check_config script gives me:

16-10-23 15:49:26 ERROR (Thread-1) [homeassistant.bootstrap] The following platforms contain invalid configuration: script (please check your configuration)expected dict for dictionary value @ data['script']['timed_light']['sequence'][0]['data']. Got [OrderedDict([('entity_id', 'light.hallway'), ('brightness', 255), ('transition', 900)]), OrderedDict([('entity_id', 'light.bedroom'), ('brightness', 255), ('transition', 900)])]. Please check the docs at

Followed by a regurgitation of my scripts with some of it in red. To be honest I’m not quite sure exactly what it is trying to tell me. Here’s a sample if anyone is interested.

I’ll keep at it.

Fixed it!

My issue stemmed from using condition: in a few of my scripts. Luckily there weren’t many and they weren’t crucial so I removed them No more async loop errors everything started working again.

I hope this helps out.

Thanks for staying on this! I got side tracked today. I will look my script file over soon. I don’t think I am using conditions in scripts but good to know it’s along those lines.

Any further info on this? I am getting the same thing every 15 seconds in my logs. I am using a few conditions in my scripts but want to keep them. The document says conditions in scripts are fine so I shouldn’t have to remove them.


I have not found a resolution. There are a lot of async pull requests so hopefully it will be resolved on the next update. I am not experiencing any performance issues, it is just annoying. It will be a long ~10 days…

Thanks for the update. Just curious, should my log entries be showing my local time instead of UTC? I have my timezone set correctly and seeing the UTC entries in the log is extremely annoying.

Just upgraded . Same for me.

I’m also noticing this issue. Is this something that we expect to be fixed in the next version, or is it a configuration error?

Same problem.

There are a handful of async pull requests on GitHub. I am imagine there will be a release on Saturday, so hopefully it is resolved. Chances are that we will see async errors and issues until the roll over is complete. It sure is annoying though.

Is anyone else that had this problem using the unofficial myq component? With .32 I haven’t had the async item in the logs but a warning about the cover.garage_door being too slow.

I had the same… mine seemed to be caused by the time date sensor. I got this error in the log:

“homeassistant.core: Error inside async loop: Exception in callback EventBus.async_fire(‘time_changed’, {‘now’:… tzinfo=)})”

… but only after about 1000 bog standard errors:

“homeassistant.core: Error inside async loop: Task exception was never retrieved”

I’ve just updated to 32.2 and it installed an additional component:

“Successfully installed async-timeout-1.0.0”

So I think it’s been addressed. So far no errors for me, but I’ll let you know if they appear again.

I know this is old but I had this issue and I found it to be an incorrectly named script inside another script. For example, I had a script on a timer that called another script. The script I was calling had a typo.

If you guys want to see the error happen in real time, start up hass through puddy so you can watch the “true” hass console. I’ve found it to be much better at debugging than the logs. To start has through the puddy:

$ sudo su -s /bin/bash homeassistant
$ source /srv/homeassistant/homeassistant_venv/bin/activate
$ hass

You can watch hass start up, and when the error occurs, look at the line above it. It will point to whats causing the error. As I said before, for me it was a miss labeled script.