Has anyone noticed that in the last couple releases if you edit a script or automation from the UI that some of them immediately display the SAVE button even though you have not changed anything?
Click SAVE, close, re-open and it’s there again…
1 Like
Sorry I don’t have a git hub account so not sure how to do that. Config_check used to be spot on but I have seen what you are talking about more recently.
It’s not a problem, it’s expected behaviour.
Load up Home Assistant in a tabbed browser. Send the tab to the background by loading another tab (any page other than Home Assistant). The backgrounded Home Assistant page disconnects after 5 min. To prevent this happening you can toggle the switch on your profile page off.
This may have been fixed and I always had that option on, or the option may have been inadvertently switched on, because I usually keep a Home Assistant tab available in my browser and since this update noticed it was often disconnected. Now that I have the option off it is not disconnecting.
I have always had the option ON… maybe that is why it refreshes when I switch back to that tab? Not a big problem for me anyway.
Try it from the other end of a geostationary satellite link with a 1.2s round trip latency 
3 Likes
Got it too, and everything upgraded fine.
Thanks for the smooth updates.
@mvn23 for the Opentherm_gw integration update!
Working well on the 5.1 firmware.
struggling with HA. can’t do any backups, message, system on freeze? just updated to 2021.3.1:
21-03-06 08:00:12 INFO (SyncWorker_7) [supervisor.docker.interface] Stopping homeassistant application
21-03-06 08:03:53 INFO (SyncWorker_7) [supervisor.docker.interface] Cleaning homeassistant application
21-03-06 08:03:54 INFO (MainThread) [supervisor.homeassistant] Update pulse/client.config: /data/tmp/homeassistant_pulse
21-03-06 08:03:54 INFO (SyncWorker_4) [supervisor.docker.homeassistant] Starting Home Assistant homeassistant/raspberrypi4-64-homeassistant with version 2021.3.2
21-03-06 08:03:54 INFO (MainThread) [supervisor.homeassistant.core] Wait until Home Assistant is ready
21-03-06 08:04:21 INFO (MainThread) [supervisor.snapshots] Found 15 snapshot files
21-03-06 08:06:55 INFO (MainThread) [supervisor.homeassistant.api] Updated Home Assistant API token
Is config check necessary before update? Isn’t it running automatically? I’ve seen couple of time when restarting HA that restart was not executed and I got message in notification center that cannot proceed because config check errors… My thought was that the same logic is applied to updates…
Configuring the AEMET OpenData integration causes other integrations such as Dark Sky to stop working.
update HA but now i got translation errors every where? any idee how to fix this
System Health
| version | core-2021.3.2 |
|---|---|
| installation_type | Home Assistant Supervised |
| dev | false |
| hassio | true |
| docker | true |
| virtualenv | false |
| python_version | 3.8.7 |
| os_name | Linux |
| os_version | 4.19.0-14-amd64 |
| arch | x86_64 |
| timezone | Europe/Amsterdam |
Home Assistant Community Store
| GitHub API | ok |
|---|---|
| Github API Calls Remaining | 4929 |
| Installed Version | 1.11.3 |
| Stage | running |
| Available Repositories | 748 |
| Installed Repositories | 10 |
Home Assistant Cloud
| logged_in | false |
|---|---|
| can_reach_cert_server | ok |
| can_reach_cloud_auth | ok |
| can_reach_cloud | ok |
Hass.io
| host_os | Debian GNU/Linux 10 (buster) |
|---|---|
| update_channel | stable |
| supervisor_version | supervisor-2021.03.4 |
| docker_version | 20.10.5 |
| disk_total | 456.5 GB |
| disk_used | 128.7 GB |
| healthy | true |
| supported | true |
| supervisor_api | ok |
| version_api | ok |
| installed_addons | File editor (5.2.0), Check Home Assistant configuration (3.6.0), CEC Scanner (2.4), Mosquitto broker (5.1.1), AirCast (3.0.0), MariaDB (2.2.2), Duck DNS (1.12.5), Node-RED (8.1.2), Google Assistant SDK (2.5.0), Git pull (7.12.1), Grafana (6.1.3), chrony (2.0.0), ADB - Android Debug Bridge (0.7.0), SQLite Web (3.0.1), InfluxDB (4.0.3), Tautulli (2.0.2), Traccar (0.11.0), Zigbee2mqtt (1.17.1-4), Home Assistant Google Drive Backup (0.103.1), Log Viewer (0.9.1), Portainer (1.4.0), FTP (4.0.1), AppDaemon 4 (0.4.4), Glances (0.11.1), Plex Media Server (2.5.2), phpMyAdmin (0.2.0), Visual Studio Code (3.1.0), Nginx Proxy Manager (0.10.0), NGINX Home Assistant SSL proxy (3.0.1), Nzbget (1.0.3), Filebrowser (2.11.0-rev3), Ubooquity (2.1.2-ls59), Transmission (dev), Sonarr (latest), Spotweb (1.0.9), VLC (0.1.2) |
Lovelace
| dashboards | 2 |
|---|---|
| resources | 0 |
| views | 1 |
| mode | storage |
Since I started to get the same error “Unknown error, see supervisor logs” I started to check not only Supervisor log (that only shows some warnings about OZWCP auto_uart being deprecated), but also home_assistant.log. To my surise I noticed that now it is flooded with life360 errors (was no suech errors prior to 2021.03):
2021-03-06 09:18:14 ERROR (MainThread) [homeassistant.helpers.event] Error while processing state change for device_tracker.life360_xxxxxxxxx
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/event.py", line 280, in _async_state_change_dispatcher
hass.async_run_hass_job(job, event)
File "/usr/src/homeassistant/homeassistant/core.py", line 435, in async_run_hass_job
hassjob.target(*args)
File "/usr/src/homeassistant/homeassistant/components/zone/trigger.py", line 60, in zone_automation_listener
from_match = condition.zone(hass, zone_state, from_s) if from_s else False
File "/usr/src/homeassistant/homeassistant/helpers/condition.py", line 668, in zone
raise ConditionErrorMessage("zone", "no zone specified")
It is a bot strange error, since it only logged for one life360 tracker, while I have few in my system. Otherwise integration works perfectly fine :- 
Another issue is with snmp sensors… these again seem to work fine (I did not found any one not working), but log is flooded by:
2021-03-06 09:31:39 WARNING (MainThread) [homeassistant.components.sensor] Updating snmp sensor took longer than the scheduled update interval 0:00:05
Would be at least nice to know which snmp sensor is causing this problem, but no explanation is added to message.
I have the same error. Seems like a dutch language thing. When you change your language to english the error is gone.
1 Like
Hmm, Nest integrations seems to not do video after update to 2021.3.x
I have only one camera, a video doorbell.
Looking in the integration it looks kinda ok:
But when I try to show the camera in the lovelace, i don’t see anything.
Please be meritorical, i dont need reading personal comments from fragile people.
Blocking it with firewall is not elegant solution.
I can’t understand why there is no option to ‘sandbox’ my working system. So i have 100% stable automation solution.
When i want change hardware (switches, plugs, bulbs, whatever) or play with new ideas i can always test it on new installation and CLOSE my new system from updating anything forever.
Just tell me, what ideology is behind this, so i cant have this CHOICE. You want stand alone install just click one button. You want new crashes, chaos or neverendig troubleshooting, ok your choice, continue using updates.
If you want standalone then you need to disconnect it from the internet
No i dont. I need OPTION to disable auto updates. I still want to connect from remote location. As i told, blocking something with firewall is from my point of view very ugly and lame solution.
Maybe someone from development team can unswer my simple question.
No updates and connected to internet is a bad idea
It’s only my problem/choice and i don’t want someone decide for me.