Hi, can’t access my Home Assistant GREEN since I upgraded to 2024.9.1.
ALL login screens on iOS app and on ALL browsers are stuck on ‘Data Loading’ after I enter my credentials. Front end never loads.
HA is definitely running because I am getting normal alerts from sensors.
HA is connected remotely according to Nabu Casa.
Tried LOCAL connection using ‘.local:8123’ - same situation.
Tried LOCAL IP address explicitly - same situation.
Tried powering down the hardware unit and powering back up - same situation.
Deleted and reinstalled the iOPS app - same situation.
Effectively LOCKED OUT of my HA even though its running…PLEASE HELP someone…
I think your issue is that the latest 2 core updates cause crashing on Sections dashboards when viewed from tablets (as far as I can tell). I can’t access my home dashboard on my iPad kiosk in the kitchen because it crashes and apparently people running android tablets are experiencing the same issue. Roll back to a previous core and you will be OK for now.
Hi Rouxdoo, thanks for your reply. How can I roll back or do anything if I’m ‘blind’ to the system? Will I need to somehow get a monitor & keyboard and connect into the HDMI port and use console? I’m not very good at line code navigation unless it’s documented clearly somewhere how to do this.
ha core update --version 2024.8.3
Please be mindful this is not a solution to the problem but only solves something for now and for you. I believe it is a good idea to search the GitHub issue and see if you can help solving the real issue. My tablets are running fine (android) so it must be something unique to you (and perhaps many others) but not a general issue.
Thank you for that. I just need to beg/steal/borrow a monitor from somewhere because I only have a laptop.
You are. Describing a very generic error. Avoid jumping to conclusions without first seeing that console screen and logs. You really need them. Get access to the logs and make sure you know how to log into togue router for your network troubleshooting while finding a way to access the physical screen of the green to see what it’s saying. Without that you’re probably gonna be stuck.
Things that may also help…
Check the router to see that the green actually has the IP address you expect and is showing active on that address.
Ping the HA box by IP from your usual laptop and make note of the results - different responses mean different things.
Ping the HA by name. For same / gives different info about name resolution.
Good tip: if you have access again, install the 'advanced SSH and web terminal ’ add-on from the community store, so you can access a HA terminal without a monitor.
The issue has been reported here and only affects people who are using the newer sections dashboard in combination with an older version of webview.
If you are not affected by this, then your webview is up to date and/or you’re not using sections.
Hi Nathan, thank you for your tips. Yes I can ping the HA Green BOTH using IP address and the domain name. I can ping the router also and I can see the HA DOES have an IP address using ‘LAN SCAN’ app on the Mac. So the Networking all looks fine. It must be a HA core/front end mess up somewhere after the update.
Thank you Shadowfist for that link, I was afraid it was just me experiencing the issue. I will try recover back to 2024.8.3 using the console…
If you have a backup from before the update (you should, because the update process prompts you to also take a backup by default), then it should simply be a matter of restoring the backup. No console needed at all.
Nevermind, I’m dumb. You still need to access the console because you can’t get into the UI. Apologies.
We all have these senior moments haha… I have a backup from a few days ago alright, if I can back in of course… othwise it’ll be a a total hard reset of the Green and a restore… never had to do that before so hopefully… …
Do you have a PC or any other non-mobile device you can use to log in?
I’m asking because I just realised that if you can’t log in from a non-mobile device (like a PC), then what you’re experiencing might not be the same issue I linked.
If you can log in from a PC, then simply restoring the backup will work. Otherwise if you can only access from a mobile, typing in ha core update --version 2024.8.3
in the console should roll you back to the previous version.
Hi, I use a Mac Laptop but all browser login attempts result in the endless spinning circle of ‘loading data’. My login credentials are fine I know. Same issue with the iOS app. I will be borrowing a HDMI monitor & a USB keyboard tomorrow so I hope all goes well with the Console CLI rollback. Worst case scenario I do have a backup a few days ago but I had been doing YAML coding editing since for a device I’m modifying that I found online so that work might be lost if I need to do a restore. Nothing major though.
If you’re using a laptop, then the issue I linked to is not what you’re experiencing. Sorry for misleading you, but when I read “iOS app and on ALL browsers”, I assumed all browsers on your mobile.
I can’t say I’ve seen this issue you reported mentioned anywhere else though. Could it be that what you’re experiencing isn’t the result of the update, and is instead caused by some YAML error from when you were modifying the device? If that’s the case, then only a restore will fix your issue.
Hi all, thanks for all your suggestions.
The only final solution was to restore a recent backup using the CLI with a terminal & keyboard.
I did manage via CLI to ‘update’ the apparently offending 91.1 back to 8.3 but that didn’t solve the issue.
I managed to ‘downgrade’ to OS 13 but that didn’t help.
Endless CORE checks/reboots/ and CLI terminal efforts to delete that “./storage” folder (as suggested in other forum articles) by doing a ‘login’ command in CLI to get into some kind of Terminal… that didn’t solve the issue… neither did endless browser cache cleans/history deletes/refreshes… nor powering down/up the HA hub.
Finally I had to do a ‘Backup Restore’ using the CLI interface … it was tricky to hunt down and find the correct backup filename… they call it a ‘slug’ reference… I do laugh at computer lingo sometimes… anyway all is restored back to a stable recent version of my system.
It was a very peculiar and dramatic issue, I had a ‘headless’ system which was still functioning perfectly.
Glad you got it sorted out!
If the rollback didn’t solve the issue, then allow me to repeat my previous question:
Hi Shadowfist, yes it would look like there were other gremlins at play… I had been doing a lot of messing about with YAML Template Trigger code for a stopwatch timer… reloading automation YAML, testing in Developers section etc… BUT it’s worrying that that could have messed up the Front End Lovelace part… I would have expected the Lovelace Dashboard to be well protected.
Remember I COULD login, my credentials were correct, the system was running AND whilst the "Loading Data’ screen was spinning away, IF I did a power down/power up of the hub, I WOULD see those little black and white messages in the lower left of the screen about ‘HA starting up’, ‘HA reconnecting’, ‘HA not fully started, some services may not…’ etc… So everything was functioning perfectly behind the scenes and messages getting through, but Lovelace dashboard could not load.
Yeah, that is indeed weird.
The good news is - you’re now in the perfect position to confirm what caused it:
- You can update Core and OS separately to see if it affects your ability to load a dashboard, plus
- You now have the knowledge to restore a backup if shit hits the fan.
It’s always the setbacks that teach us the most.