Matter server add-on DNSSD errors

I added the matter server add-on and when I run it these 2 errors are flooding the log

2023-02-04 01:31:46 core-matter-server chip.DIS[123] ERROR DNSSD packet parsing failed (for SRV records)
2023-02-04 01:31:46 core-matter-server chip.DIS[123] ERROR DNSSD packet parsing failed (for non-srv records)

Any thoughts on what could be causing it? mDNS seems to be fine on the network as I use the Air cast add-on that find my Chromecast just fine.


I recall something about ZigBee using IPv6, so maybe you have issues with IPv6 on your network.

Only IPv4, no IPv6 on the network.

Same issue here. IP4 Only. HAOS. Running the SkyConnect USB dongle.

Home Assistant 2023.3.1
Supervisor 2023.01.1
Operating System 9.5

2023-03-07 07:55:02 core-matter-server chip.DIS[126] ERROR DNSSD packet parsing failed (for non-srv records)
2023-03-07 07:55:02 core-matter-server chip.DIS[126] ERROR DNSSD packet parsing failed (for SRV records)

Any luck figuring it out?

When I posted about IPv6, I meant that it does not use IPv4 at all, so a faulty IPv6 setup would affect the services.

I see the same. I am also using ipv4 however ipv6 is setup and working, HassOS has an ipv6 address in the network section.

I have the same issue and have been unable to add a nanoleaf essentials bulb via matter. I wonder if this is the reason why.

It’s likely not the issue, I managed to get my matter device, a Tapo Smart plug, added and continue to see this error in the log. Have you added the matter device to either Google/ Alexa / Homekit? I added to Google Home first and shared from there to HA.

I still have this problem too. Could be that I’m using a CM4 with a wireless interface on my Home Assistant Yellow? It clogs up the logs and makes it harder to troubleshoot.

I stopped using the matter device for now, not stable enough for me.

Same issue here

Are any of you using multiple LANs/VLANs with multiple network interfaces on the HA host?
If so then you might need to configure your routes.

How would you configure them?

I have been trying to wrestle the IPv6 beast a lot lately and if you are using multiple network interfaces on HA, then you are in for a fight of the big ones and your network understanding from IPv4 will not work at all. IPv6 is so different.

What you need to configure is the network binding of the services and I have no clue where to do that.
In IPv6 every network interface will have several networks, like fe80::/64, ff02::/16 and so on.
If you have 2 network interfaces, then it means you will have 2x fe80::/64, 2x ff02::16 and so on.
When you then want to contact something on a fe80::/64 network then you need to specify which network interface this is going to happen on. With just one network the operating system will always choose the right one, but with multiple it will choice the one that it finds appropriate if nothing is stated.
This might not be the right one, so the standard have a solution to that, which is to state the network interface ID in the address. The interface ID is added to the end with %ID_number_of_interface.

IPv6 will also add multiple addresses that act differently, like some that will not be routed between networks/VLANs, like fe80, others that can be routed between local networks/VLANs, but not on the internet, like fc00::/10.
And these addresses can exist in stateful versions, where it is controlled totally by a server/router, in a stateless versions, that is controlled partly (the prefix part) by a server/router, also called SLAAC, and also in temporary versions that is handled without any server/router.