-
-
Notifications
You must be signed in to change notification settings - Fork 31.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Hassio does not load correctly #21774
Comments
As my research in #21619
It looks like I got the conclusion too soon for your case. I will look at it tonight. If possible, @daenny could you check if you have a log entry for hassio begin to setup? How long it take till the CancelledError occurred. |
Could you share your full DEBUG level log to me? You can ping me at discord channel. |
Hi, thanks for looking into it. I see the warning the setup of Hassio is taking longer than 10 seconds. I can have a look if I can get the debug logs tonight. Any components in particular or the complete debug log? |
I feel that something special on your system caused dead lock in the hassio component setup. I opened a pr #21794 to fix the auth_store race condition, but I am not sure this will fix you root cause, so a full debug log will be helpful. |
This looks similar to #21811. 19-03-08 17:46:10 INFO (SyncWorker_16) [hassio.docker.homeassistant] Start homeassistant homeassistant/raspberrypi3-homeassistant with version 0.89.1 |
@ghitchen they are totally different |
Same problem for me. |
@awarecan sadly I think this does not solve my problem. To test the changes, I did the following:
I also disabled the mosquitto broker and node-red plugin for testing. Since I though that those might be the problem as I think it heavily query the auth_store (I use authenticated mosquitto messages) |
We only load auth_store from file system once. Write it back every minute if there are any changes. What is your current error message? |
Same error. CancelledError. |
I think the CancelledError might be caused by the hassio component not be able to finish setup in a long time. Maybe because your zwave network is starting up at the same time. zwave writes very verbose debug log, which will slow down the already struggled system more. |
I pinged you on discord, my zwave network is very small, only 4 nodes. On a successful start I see a lot of these messages:
The IP is of the supervisor. They are multiple per seconds. |
Behaviour is better for me, when I disable the plugins of node-red and zigbee2mqtt. I start them with an automation 60s after HA started and stop them on shutdown. It still does not always work but at least success rate of restarts is now at ca 50%:
|
@daenny do you have |
Alright, I think I found the root cause. There are several issues actually, but in short, put
Another way is to increase the size of thread pool. You should have 10 threads in your HA's task execution pool. I guess you will get rid of the problem by increase executor_opts = {'max_workers': 20} # type: Dict[str, Any] |
You can also disable the problematic Miflora sensor. |
Thanks for you deep insights. I am currently away from my system until the weekend. It's currently running, so I will keep it that way until I am back. I will when I am back with your settings, i.e. adding EDIT: And as I have a dual-core CPU I have 10 threads, as I am using 2 sensors, those indeed will completely exhaust all available threads... |
No problem. I will have a look tonight. |
Just ignore it, my PR has a problem, need redo. |
Ok. Adding ’hassio:’ to the config resolves the problem of not booting up. With that I can reliable restart even with zigbee2mqtt and node red enabled. I also updated to v0.90.1. |
Home Assistant release with the issue:
0.89.0
Last working Home Assistant release (if known):
0.88.2 (partly)
Operating environment (Hass.io/Docker/Windows/etc.):
Hassio on Ubuntu host
Component/platform:
Maybe related to:
#21569
Description of problem:
Hassio frontend does not come up. Even multiple restarts do not solve it.
Problem-relevant
configuration.yaml
entries and (fill out even if it seems unimportant):Traceback (if applicable):
I have a automation which restarts nodered add-on 60 seconds after home_assistant start. This is also not done correctly.
Additional information:
The text was updated successfully, but these errors were encountered: