Local IP address not working but hostname.Local working

Configuration
1

Hi everyone, :smiley:

I’m having an issue in my Home Assistant OS running on nuc. Where I gets stuck on the authentication screen when accessed via its local IP address: http://192.168.68.74:8123. It remains stuck at the following URL:

(http://192.168.68.74:8123/auth/authorize?response_type=code&redirect_uri=http%3A%2F%2F192.168.68.74%3A8123%2F%3Fauth_callback%3D1&client_id=http%3A%2F%2F192.168.68.74%3A8123%2F&state=eyJoYXNzVXJsIjoiaHR0cDovLzE5Mi4xNjguNjguNzQ6ODEyMyIsImNsaWVudElkIjoiaHR0cDovLzE5Mi4xNjguNjguNzQ6ODEyMy8ifQ%3D%3D)

The same instance works perfectly when accessed via the .local hostname (http://homeassistant.local:8123) or via the external URL through Nabu Casa.

Steps I’ve Taken to Resolve the Issue

  1. Cleared Browser Cache:
  • Tried accessing HA in incognito/private mode and cleared the browser cache. The issue persists across multiple browsers (Chrome, Edge, Safari) and devices.
  1. Checked Network Configuration:
  • Verified that the local IP address (192.168.68.74) is correct and reachable. I can ping it successfully from other devices.
  • No firewall is blocking access, and all devices are on the same local network.
  1. Safe Mode Restart:
  • Restarted HA in safe mode to disable all custom integrations and components. However, the issue persists even in safe mode.

5.Reviewed Logs:

  • Checked HA logs but didn’t find anything directly related to this issue.

  • I noticed unrelated warnings from custom integrations like tplink_easy_smart, frigate, and moonraker, but these integrations were disabled in safe mode.

  • Rebuilt HA Core:

  • Rebuilt HA core using the command ha core rebuild, but the problem remains.

  • Checked and Configured Internal URL:

  • Verified that the internal_url in configuration.yaml is set correctly:

homeassistant:
  internal_url: http://192.168.68.74:8123

Current State

I’ve exhausted the troubleshooting steps I could find. HA remains stuck at the /auth/authorize page when accessed via http://192.168.68.74:8123. This issue started after I restored HA from a full backup.

Questions for the Community

  1. Has anyone encountered a similar issue where HA gets stuck at the /auth/authorize page via a local IP address?
  2. Could this be related to the TP-Link switch or Deco X60 configuration (e.g., QoS or multicast settings)?
  3. Are there any additional debugging steps or configurations I should try to resolve this?

Any insights or guidance would be greatly appreciated. Thank you in advance for your help!

2

Hello, i had same issue on a VM oracle.
Not sure if it will help you, but have a look: Installation issues - #12 by DocKoala992

Also, have you reserved the port in the router? - i know you can access via localhost, but is it reserved?

1 Like
3

thank you for your replay
1/ ipv6 disabled - tested same result
the rest apply for setting up the network for the VM which not applied in my case
as I’m running HA OS directly on the NUC

2/Yes the IP is reserved for the HA OS from router

4

What IP is resolved when you ping homeassistant.local ?

Have you tried clearing your browsers cache (or tried incognito/inprivate mode)?

2 Likes
5

Hi thank you for your replay :smiley:

What IP is resolved when you ping nuchomeassistant.local ?

Ping statistics for fe80::416d:ca76:f70:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 0ms, Maximum = 0ms, Average = 0ms

Have you tried clearing your browsers cache (or tried incognito/inprivate mode)?

Yes tried both and no luck the same issue , loading and stuck on loading

1 Like
6

Ok that means the difference is when you use by name you’re getting name resolution and connecting by IPV6 (the fe80 is a link local ipv6)

But when you connect using the IPV4 address you get something different.

Something is jacked up on the IPV4 side of your network. Most ping commands have a method to force ping over IPV4. Do ping /? Figure out that command and run it against the same host. It should try an IPV4 address. Let us know what happens

1 Like
7

Try with
ping -4 homeassistant.local
(-4 forces TCPIP4 on windows) :wink:

2 Likes
8

I try it and the results :

Pinging nuchomeassistant.local [192.168.68.74] with 32 bytes of data:
Reply from 192.168.68.74: bytes=32 time<1ms TTL=63
Reply from 192.168.68.74: bytes=32 time<1ms TTL=63
Reply from 192.168.68.74: bytes=32 time=1ms TTL=63
Reply from 192.168.68.74: bytes=32 time<1ms TTL=63

Ping statistics for 192.168.68.74:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 0ms, Maximum = 1ms, Average = 0ms
9

Hi, thanks for the suggestion! Here are the steps I followed to test the ping for both IPv4 and hostname:

  1. Ping by Hostname (nuchomeassistant.local):
ping nuchomeassistant.local
  • This successfully resolved to the IPv6 address (fe80::xxxx) and returned responses without any issues.
  1. Ping by IPv4 Address (192.168.68.74):
  • I forced the ping to use IPv4 with:
ping -4 192.168.68.74
  • The ping to the IPv4 address also worked perfectly, with consistent replies.
  1. Result:
  • Both the hostname and IPv4 address are reachable, and there doesn’t appear to be any connectivity issue at the network level.

Since the pings are working fine, it seems the issue might be more related to how Home Assistant handles requests via the local IPv4 address. Let me know if you have any further suggestions or insights into what could be causing the issue. Thanks again for your help!