I’m having trouble accessing my ha via the web at homeassistant.local.8123
I can access my ha through the ha ip.
I think it has something to do with the ip address not being assigned properly. I can’t see my ha ip on my router’s network map.
Anyone be able to help troubleshoot?
it should be homeassistant.local:8123
Note the : (colon) ASCII 58
If you do not have an SSL/TLS certificate loaded you may have to prefix that with http://, instead of the default https://
I have tried with the correct homeassistant.local:8123, the two periods was a typo on my part!
Is your Home assistant implementation called “homeassistant”?
Asking, because in general .local domains in this instance would be using mDNS (RFC6762), which means it would be [machine name].local.
Look under Settings → Network → Host name: The name your instance will have on your network.
For example, one of the HA instances I run it would be hakaishin.local not homeassistant.local
Yes it is named homeassistant
So, to recap, you can access the HA through IPv4 address, but not through homeassistant.local. Correct?
What is HA running on?
Asking because this might be a mDNS problem with the host that is running HA.
Is HA a static IP, or dynamic?
Do you have the right IP? (try local_ip integration)
What do you get when you ping -4 homeassistant.local?
If that errors out, what does ping -6 homeassistant.local show?
Is there a firewall, switch, IPS that maybe blocking multicast?
Do you have any other “named” devices (e.g., router) other than your workstation? If so can you ping them with ping -a <machine_name.local>?
Finally, can you nmap -p 5353 <HA's IP>? Alternatively you can try a dig command like
dig +nocmd +noquestion +noauthority +nostats _http._tcp.<HA's IP> - these will attempt to see if mDNS is running on HA.
I do not know what is your workstation OS, so the commands might be different.
Are you on the same network?
Check if your workdtation IP address is the same as your HA server IP address in the places where you subnet mask is named 255.
My HA is running on an intel nuc, has a dynamic ip.
When I try to ping homeassistant.local on both ipv4 and 6 i get “Ping request could not find host homeassistant.local. Please check the name and try again.”
I don’t believe there is anything that would block it, it was working a bit ago before I got a new router.
Yes, I can ping my NAS that is named with the command
Nmap command gives me this
Nmap scan report for 192.168.1.66
Host is up (0.00015s latency).
PORT STATE SERVICE
5353/tcp closed mdns
I think it may have something to do with getting a new router? The gateway on the router is 192.168.50.XX while the HA ip is 192.168.1.XX. Would that be causing some issue?
Could be.
What is the IP netmask for your workstation that you are trying to reach HA?
Is the DHCPd on the new router?
Is it handing out the same netmask?
on windows ipconfig /all and look for subnet mask, in Linux ifconfig, or ip a or ip -d all will show the subnet mask aka prefix length.
Maybe before it was 192.168.1.0/24, but now it is 192.168.50.0/24?
mDNS is not made to cross subnets. It is possible that your router happily passes on the IP address, but the mDNS name (homeassistant.local) just gets dropped.
Netmask is 255.255.255.0
In the router LAN setting the subnet mask is set to the same
Is there a way to reset the HA so it gets a new 192.168.50.xx ip from the router? Or is it better if I try to set a static ip for it?
Is that going to help or is there something else going on?
If HA is using a dynamic IP then restart it and it should all be back with an IP address in the correct network.
Yes that could work, if the DHCPd did not cache the IP because the lease is infinite or just very long.
I have tried restarting but no changes. My HA does have 2 DNS servers in the automatic network settings. Would that change things, should I delete them?
DHCPd runs on the router and you have changed that one.
DHCPcd runs on a Linux client and does not cache an IP address.
Okay, so any thoughts on what I could try?
Restart the HA host.
Preferably through the console, but if need be it can be done by pulling the power.
This. If it’s truly DHCP it WILL get a new address within subnet.
DHCP clients check themselves on reboot. Then start trying to renew thier address at half lease remaining. If you had a common consumer subnet (8-24 -hrs) that time has already passed for you. (reading age of issue)
If it does not get a new IP. Something is giving it its old IP. Or it’s static.
