Can't reach gui at :8123 anymore

Installation
1

Perhaps somebody can give me a hint…

I’m running HA in Proxmox VM. VM has static IP. HA is working fine and performing all its tasks. All of the sudden I can’t reach the interface via port 8123 anymore (perhaps installing 2024.12.4 did something?).

From hypervisor I can reach the observer via port 4357, can SSH via port 22, no problem. 8123 is blocked all of the sudden. Firewall on Proxmox is turned off for testing.

root@pve:~# telnet 192.168.1.99 8123

Trying 192.168.1.99…

telnet: Unable to connect to remote host: Connection refused

Restoring backup from 2 days ago is not solving the problem…?? While than all was working fine. Strange to me.

Any hint?

2

Since you are running under Proxmox, open the VM in Proxmox, select Console. At the ha prompt type login and press enter.

At the linux prompt enter docker ps and press enter.

Do you see the homeassistant docker container running? It usually ends in the version number

If you enter the command

netstat -tunlp | grep 8123

Is there any output?

3

Dear Martin,

tnx for you reply and your suggestions.

Yes, homeassistant docker is running. It also function properly etc, responding via port 4357.

And no, netstat -tunlp | grep 8123 returning only prompt.

ha
ha1013×593 62.3 KB

4

The observer web service (4357) is a separate docker container (the last one in your list). Since the home assistant docker process is running but the web server isn’t, you probably have something wrong in configuration.yaml. I would check the log file or restart in safe mode. Were you making any configuration changes before it failed?
command line safe mode:

ha core restart --safe-mode

5

Dear Martin,

Once again thank you for your input.

You are absolutely right, observer is just another container. My bad. Learning thou, thanks a lot for that.

I didn’t change anything manually in configuration files.

The changes that took place lately is HA core 2024.12.4 and Proxmox VE went from 8.3.1 to 8.3.2.

I did try already (and now again) restarting core in safe mode with no success.

Now the new part: I thought to keep the problematic VM image for further investigation but for the time being, built new HA with help of the new Proxmox functionality, importing .ova (worked great) and restoring few days old backup. It worked. Very well, although few days of history got lost of course.

Did some upgrades to core 2024.12.5 and it worked for few hours… and then… the same happened again. HA functionalities are working well but no access via port 8123. Wow…

Now I’m starting suspecting Proxmox and its networking virtual adapter.

In the newly installed HA after attempt of changing IP from DHCP to static, while changing IP number, network mask go changed automatically to 255.255.255.255 and locking me out.

That was while attempting IP change manually. Perhaps this sort of mask change take place randomly… that would explain a lot.

Now my question: in HA → settings → system → network, option can be found: network adapter ‘auto configure’. Should I try to switch this option off?

Or should I look more in the direction of network adapter within Proxmox VE. Now I have “VirtIO (paravirtualized)”. Same one that I have already for long time.

Long story, sorry for that.

6

I would check the network settings in the CLI. Personally, I doubt it has anything to do with Proxmox or network settings.

What does this mean? HA functionalities are working well?

No access makes sense if no process is listening on port 8123. You established that that there is no web server on port 8123. The question is, why is HA not starting the front-end web server successfully? Seems your problem is more software related.

7

Tnx again.

I’m going to read myself a bit in ‘check network settings in the CLI’ since I would’t know what I could do.

About the functionalities I meant all the automations, messages that HA sends to me ect. That all works. Indeed reaching web serwer is impossible, no wonder since there is nothing listening at 8123.

8

Checking network settings from CLI didn’t bring me any further.

After few attempts of restoring from backups and installing updates core 2024.12.5, supervisor 2024.12.0 and OS 14.1 got the system stable again… for few days.
Then all of the sudden Z2M gave up. Restarting addon / container didn’t help. Restoring from backup also not. There was no communication with Z2M container possible. What did helped was rebooting Proxmox. After rebooting, now the whole system is on for two days again, with no problems. Hmmm…
I’m running HA supervised with all the add-ons in on VM.