I didn’t want to have to make a post here, but I have spent almost the entire week trying to solve my issues. So my setup is:
Unifi Dream Machine SE
Two APs, one hardwired one in mesh mode in the garage
Proxmox PC running HomeAssistant in a VM
Multiple VLANs
Proxmox on server VLAN with HA tagged to the IoT VLAN
Matter server add-on installed and the Matter integration
Linkind Matter bulbs
I am able to get the bulbs to connect via the companion app and work for a little while. They will have great performance, then at some random point in the next few hours they decide to break, not work consistently, and just be weird.
I get multiple mDNS timeouts and failure to establish CASE errors in the Matter server logs. I have enabled ipv6 on the network, RA is on high, Proxmox Multicast snooping is off, I have mDNS enabled on the network, no IGMP snooping or multicast-unicast controls. I cannot for the life of me figure out why it won’t work consistently. Please help, I just want to enjoy my cheap Matter devices, unless you guys thing I should just move to Zigbee.
Why all the VLANs?
How are you making sure the devices connect (and more importantly re-connect) to their correct WiFi IP Address? Consistently? IPv4 vs IPv6?
how many iot ssid do you have ? some wifi iot device doesn’t like roaming enabled ssid (802.11k/v/r). at the very least check whether your iot device still connected to your net
I have them so I can separate things like my personal network from IoT, Unifi devices, guest, and my servers/services. I haven’t had any issues with them until installing home assistant and trying to use Matter. In unifi it shows that the linkind bulbs are only getting ipv4 addresses although other devices on the same VLAN are getting ipv6 addresses and are pingable (HA install and Google Nest Hub). Home assistant’s matter integration however shows the bulbs as having ipv6 addresses when connected. I have ipv6 setup only locally now with it being static with SLAAC. I dont have anything setup that forces the bulbs or devices to keep the same address.
One VLAN and one SSID for IoT devices. The devices on unifi are still available and show as being up/online. First ping does fail and say host is unreachable, then the next ones work and it pings back but then on another device it said host unreachable the entire time. I swear I am just breaking things more each time I touch it
I read something about the timing with matter being very critical when not directly via USB connected. The communication seems to be a simple serial protocol without error checking. So 1 bit gets screwed, you’ve got a problem.
So I think you have to search in the LAN/WIFI/VLAN NETWORK and/or the CPU performance of the router.
