I have adguard installed so I'm thinking it has something Todo with it. But I'm unsure of wat the domain .local or .lan is referred as so my googling was not very helpful.
Anyway the issue I'm having is on the status page it shows the active DHCP leases and shows their hostname and inside () shows the hostname.local. but that last one keeps disappearing and then re-apearing randomly.
Any thoughts?
.local is the reserved domain for mdns (auto-discovery/zero configuration). Devices will register themselves automatically with their respective hostname.local, but this will come and go if the device is disconnected from the network (including sleeping/off).
The .lan domain is somewhat arbitrary — it is the default domain for your LAN on OpenWrt, but can be changed if you have a real domain name you’d rather use (or there are a few other proposed/commonly used options for non-internet/private domains intended for use in home networks). These will populate based on the DHCP request for a device, if that device provides a value in the hostname field. Also, if you set a dns record (often when you set a DHCP reservation), the hostname.lan should appear in that context as well.
I changed the default domain domain to .local since chrome doesn't route .lan locally. So I just set it to .local instead.
And about the sleeping/disconnecting possibility, would that be happening every few seconds to a min? I was thinking the router was resetting it and then adguard was adding it back or something. Since before I set adguard to listen to all interfaces, the .local domain wasn't even showing up in the router just adguard.
Probably a bad idea because it clearly violates standard. Having .lan is better but it is not standardized too. As far as I understand we have .home.arpa and .internal as our choice from ICANN.
Really? I was just taking the example from home assistant which uses .local. that surely wouldn't be the cause for the router page constantly removing them re-adding the domain to the list would it?
Yes... in the end you are in control and chose whatever pleases you. Yet .local has the side effect of clashing with on-by-default mDNS/bonjour. I would also try to avoid reusing an internet wide active top level domain, like .com or .org, outside of these you can pragmatically pick anything (just be aware that ICANN can decide to introduce new top level domains when ever they feel like it).
This is how mine looks. Those .local in the ()s however constantly disappear and then come back within a few seconds or 10. Besides that everything seems to work fine. It's just annoying to see the info randomly resetting.
Does this also happen, if you, for testing purposes, switch your local dmain name to, say, .lokal with a K? My gut feeling is what you observe is the side effect of using .local as local domain name and having mDNS active...
So, can you use hostnames like host.lan or not?
Return to default .lan config and start from very beginning. Most likely you have some faulty embedded client (if problem exists at all). Don't forget to reboot your clients after that or you may have to wait up to 12 hours until lease time is expired and those clients will appear in status page.
And see System log. Maybe you're spammed by DHCPDISCOVER-DHCPOFFER-DHCPREQUEST-DHCPACK sequences. I remember I had Dlink powerline adapter which requested new address every minute.
I will try setting the hostname back to .lan and see if it still does it. Im a little busy today so I may just wait out the dhcp lease after I make the change instead of going thru and resetting devices.
Ok so I didn't even have to wait an hour it seems. The issue was the same. The status page keeps adding then removing the "(device.lan)" after the hostname of the device. I don't seem to see a pattern.
What do your AdGuard Home upstreams look like?
How is AdGuard Home and dnsmasq configured? What is listening on which port?
Upstream servers
(3 doh servers)
[/Pool.ntp.org/]1.1.1.1
[/Local/]127.0.0.1:54
[//]127.0.0.1:54
Fallback
1.1.1.1
Bootstrap(left at default)
9.9.9.10
149.112.112.10
2620:fe::10
2620:fe::fe:10
Private reverse DNS
127.0.0.1:54
And the router DNS port is set to 54. I believe I left everything else alone.
I do have each interface(including wan) set to use custom DNS on 192.168.1.1 and DHCP options 3 and 6 set to that IP as well.
Remove [//]127.0.0.1:54 from upstreams.
Add [/168.192.in-addr.arpa/]127.0.0.1:54 to private reverse DNS.
Optional, but remove fallback.
I don't think you should set internal DNS to the WAN interface.
Ok il try those wen I get home.
Ok so onset then last night and as of right now I'm still seeing them come and go. I'm really starting to suspect it might just be a bug with openwrt and it's use with adguard. Since I'm not having connection issues or anything when using those .local domains. So it might just be a visual bug in openwrt?
The other thing I might try is enabling DHCP enhanced logging and see if the logs show an excessive amount of calls but personally I don't think it's a single or multiple devices since they all have the same visual 'bug' within the router status.
OK, I think you need to share your OpenWrt-side configs, since we eliminated the AGH upstreams for now.
Remember to redact passwords, MAC addresses and any public IP addresses you may have:
ubus call system board
cat /etc/config/dhcp
cat /etc/config/firewall
cat /etc/config/network