On 7/4/2024 at 3:34 PM, xee said:

Unfortunately no, I also tried the beta of Unraid 7 once but unfortunately this did not work either. I can't get unraid itself or Docker running on Unraid to connect to the Matter devices. I have now accepted that it only works in a VM. That's ok for now and you can live with the workaround.

 

Have you tried to set (see https://github.com/home-assistant/core/issues/115361)

 

sysctl -w net.ipv6.conf.wlan0.accept_ra=1
sysctl -w net.ipv6.conf.wlan0.accept_ra_rt_info_max_plen=64

 

Instead of wlan0 you probably need set eth0 or br0 (or both) to those values. And if that still not helps, try ra=2.

 

EDIT: I can confirm that "ra=2" has helped to add the devices by docker. My interface was br0.

Edited July 9, 20241 yr by fir3drag0n
solution

Yes, I can confirm that the two commands also solved the problem for me. Now I am only faced with the challenge of how to persist these commands so that the setting is retained after a reboot. Unfortunately my Linux knowledge is not so good.

On 7/8/2024 at 5:01 PM, fir3drag0n said:

 

Have you tried to set (see https://github.com/home-assistant/core/issues/115361)

 

sysctl -w net.ipv6.conf.wlan0.accept_ra=1
sysctl -w net.ipv6.conf.wlan0.accept_ra_rt_info_max_plen=64

 

Instead of wlan0 you probably need set eth0 or br0 (or both) to those values. And if that still not helps, try ra=2.

 

EDIT: I can confirm that "ra=2" has helped to add the devices by docker. My interface was br0.

this need to run on the host unraid? is this a terminal command?
User script plugin and run the script at startup of array.

Is this a VM? Use crontab or rc.local

rc.local: https://www.cyberciti.biz/faq/how-to-enable-rc-local-shell-script-on-systemd-while-booting-linux-system/

crontab -e @reboot boot.sh ... with the boot.sh contains the needed terminal commands... https://www.linode.com/docs/guides/run-jobs-or-scripts-using-crontab-on-boot/

 

Docker? a bit trickery as you would need to add to an environment and or edit the docker code to add that to boot
unraid user script to execute command when docker running would need to run sleep command for 120 sec (2 Min) to make sure it is on for command to execute at startup of array...

 

https://stackoverflow.com/questions/47671589/how-can-i-run-script-automatically-after-docker-container-startup

https://docs.docker.com/reference/cli/docker/container/exec/
EX for docker: docker exec -it -u root PhotoPrism /bin/bash
^ puts you in root terminal of a docker names PhotoPrism docker... /bin/bash is executed. one could put a file in via template to execute a script for example...

Hope this helps with automation to make a per script for your needs.

The tip with the “User Scrip” plugin is great, it worked immediately for me. Now the Matter Server is fully operational and I was able to deactivate the VM and now have everything on Docker.

 

Many thanks to you all.

On 7/9/2024 at 6:01 AM, fir3drag0n said:

 

Have you tried to set (see https://github.com/home-assistant/core/issues/115361)

 

sysctl -w net.ipv6.conf.wlan0.accept_ra=1
sysctl -w net.ipv6.conf.wlan0.accept_ra_rt_info_max_plen=64

 

Instead of wlan0 you probably need set eth0 or br0 (or both) to those values. And if that still not helps, try ra=2.

 

EDIT: I can confirm that "ra=2" has helped to add the devices by docker. My interface was br0.

Thanks for that first, I haven't verify above setting strictly necessary or not, because I set that still fail to pair matter device. After realize Matter server must in host network then every thing start to work.

 

Just for sharing , below are the problem

- Matter server ( CA docker ) must in host network and Matter device also need in same network, but people usually set IoT device in different network.

- HA in custom network can connect Matter server which in host network, but this need IPv4 routing between them and enable host access.

- Matter device usually only support limited function, for example Nest thermostat have humidity sensor, but Matter won't support that, and as I know, smart plug power metering also not support.

 

 

 

Matter already develop so many year still useless, really disappointed !!

Edited September 11, 20241 yr by Vr2Io

I'm trying to set this up as well but not having as much luck and not sure what to do with now.

Matter server on host networking
HA on custom network br0

I have also set up unraid docker with macvlan
I can ping between the matter-server and HA docker
 

I have also done

sysctl -w net.ipv6.conf.br0.accept_ra=2
and
sysctl -w net.ipv6.conf.br0.accept_ra_rt_info_max_plen=64

When I run
 

ip -6 route

::1 dev lo proto kernel metric 256 pref medium
fde1:4182:c0cf:1e41::/64 dev br0 proto kernel metric 256 expires 1668sec pref medium
fde1:4182:c0cf:1e41::/64 dev br0 proto ra metric 1062 pref medium
fdf4:ed3b:27bd::/64 via fe80::10c0:c750:e4e:c6b2 dev br0 proto ra metric 1024 expires 1668sec pref medium
fdf4:ed3b:27bd::/64 via fe80::255:daff:fe5f:ce80 dev br0 proto ra metric 1024 expires 1696sec pref medium
fdf4:ed3b:27bd::/64 via fe80::1436:46b9:d7e0:4b62 dev br0 proto ra metric 1024 expires 1630sec pref medium
fdf4:ed3b:27bd::/64 via fe80::469:4336:3564:9cb0 dev br0 proto ra metric 1024 expires 1652sec pref medium
fe80::/64 dev br0 proto kernel metric 256 pref medium
fe80::/64 dev bond0 proto kernel metric 256 pref medium
fe80::/64 dev vetheb1a7f9 proto kernel metric 256 pref medium
fe80::/64 dev veth704bd15 proto kernel metric 256 pref medium
fe80::/64 dev veth8519fa4 proto kernel metric 256 pref medium
fe80::/64 dev veth8c1d9a3 proto kernel metric 256 pref medium
fe80::/64 dev veth9b51bd1 proto kernel metric 256 pref medium
fe80::/64 dev veth4fd48ba proto kernel metric 256 pref medium
fe80::/64 dev vethf8c79e6 proto kernel metric 256 pref medium
fe80::/64 dev veth4749be0 proto kernel metric 256 pref medium
fe80::/64 dev veth3d8804a proto kernel metric 256 pref medium

 

Then when I try to add a device that is already provisioned I get chip error outputs. It does seem to say that it can establish a connection.

 

Any help would be appreciated

 

update: I was able to get it to work now.

 

I didn't have a ipv6 gateway configured by my router so I had to manually create one.

 

Note the gateway address I created 

fe80::1

once I did this along with the br0 configuration of ra=2 I can see a bunch of ipv6 addresses of my home thread devices appear in my ip -6 route

6 hours ago, Mongo said:

update: I was able to get it to work now.

 

I didn't have a ipv6 gateway configured by my router so I had to manually create one.

 

Note the gateway address I created 

fe80::1

once I did this along with the br0 configuration of ra=2 I can see a bunch of ipv6 addresses of my home thread devices appear in my ip -6 route

 

 

👍👍I never touch / add any IPv6 setting in Unraid, just enable IPv6 only in Unraid, physical network still IPv4 only.

On 7/9/2024 at 12:01 AM, fir3drag0n said:

 

Have you tried to set (see https://github.com/home-assistant/core/issues/115361)

 

sysctl -w net.ipv6.conf.wlan0.accept_ra=1
sysctl -w net.ipv6.conf.wlan0.accept_ra_rt_info_max_plen=64

 

Instead of wlan0 you probably need set eth0 or br0 (or both) to those values. And if that still not helps, try ra=2.

 

EDIT: I can confirm that "ra=2" has helped to add the devices by docker. My interface was br0.

Hi,

 

can you share your routing table please ? It's doesn't work for me

9 hours ago, JamesAdams said:

Hi,

 

can you share your routing table please ? It's doesn't work for me

did you change wlan0 to (br0, bond0 eth0) for your interface name?

make a separate forum post and attached a diag file...

did you enable ipv6 forwarding?

 

24 minutes ago, bmartino1 said:

did you change wlan0 to (br0, bond0 eth0) for your interface name?

make a separate forum post and attached a diag file...

did you enable ipv6 forwarding?

 

 

Hi, thanks but already try and nop doesn't work

 

My topic : 

 

I'm running on a QNAP NAS and the below was a slam dunk to get the docker container to work.  Even in HAOS VM with the matter addon, it wouldn't work.

 

I added these to the user scripts on array start.  

 

sysctl -w net.ipv6.conf.br0.accept_ra=2

sysctl -w net.ipv6.conf.eth0.accept_ra=2

 

The device was added instantly when trying to pair, as compared to previous failures.