-
Notifications
You must be signed in to change notification settings - Fork 43
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update README.md with info that Avahi daemon and Chrome on Linux should be stopped #68
Comments
FYI: Probably, you need to change $ sudo ss --listen --all --processes --num|grep 5353
udp UNCONN 0 0 0.0.0.0:5353 0.0.0.0:* users:(("systemd-resolve",pid=660,fd=11))
udp UNCONN 0 0 [::]:5353 [::]:* users:(("systemd-resolve",pid=660,fd=12)) |
I came to this after running into issues with Additionally, stopping avahi is not as trivial as |
Which is why we also have #67
Yes, this can go into the readme. |
@ivmarkov what are your thoughts on adding a |
If somebody (you?) want to spend time on making it work - by all means do. I was unable to find a Rust crate that can talk to Avahi. IMO, for real (embedded) use cases, this does not bring much value, as you want a tightly controlled, minimal Linux stack, possibly created with |
Of course, in my case I'm looking at implementing a Matter bridge on my home server for a non-Matter set of devices so I'm not really resource constrained but I also don't want to prevent other processes from making use of mDNS at the same time. The zeroconf library claims to support avahi and depends on |
Maybe I misunderstood this PR but as far as I can tell this didn't fix the issue for me. Still panics without disabling the demon on Ubuntu 22.04: [2023-10-06T20:04:44Z INFO rs_matter::transport::core] Transport: waiting for incoming packets as core::future::future::Future>::poll Steps to reproduce: |
I would recommend copying the example code into an application and then initializing with the zeroconf feature. Sidenote: @kedars what do you think about splitting the examples on a per platform basis? There's a lot of extra noise in the example in making the same source file support esp32 and Linux. |
I just on the master commit 3d8fc59, also met this issue. On arch linux. [2024-03-21T08:15:21Z INFO onoff_light] Will use network interface ens18 with 172.16.10.109/fe80::10bd:11ff:fef8:35d5 for mDNS
Error: Error::StdIoError: Address already in use (os error 98) {
disabled backtrace} |
This issue is no longer relevant because the current |
On mine node, sudo lsof -i -P -n | grep 5353
systemd-r 457 systemd-resolve 15u IPv4 18562 0t0 UDP *:5353
systemd-r 457 systemd-resolve 16u IPv6 18563 0t0 UDP *:5353
cargo build --features async-io,zeroconf --examples
And the error log is that:
2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key D val 250
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key CM val 1
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key DN val OnOff Light
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key VP val 65521+32768
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key SII val 5000
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key SAI val 300
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key PH val 33
[2024-03-21T08:28:35Z INFO rs_matter::mdns::builtin] mDNS TXT key PI val
[2024-03-21T08:28:35Z DEBUG zeroconf::linux::service] Registering service: AvahiMdnsService { client: None, poll: None, context: AvahiServiceContext { name: Some("135098307ADC8921"), kind: "_matterc._udp", port: 5540, group: None } }
[2024-03-21T08:28:35Z ERROR rs_matter::mdns::builtin] Encountered error registering mDNS service: could not initialize AvahiClient
|
Should i also need disable it with MulticastDNS=no? Now I just disable mdns in systemd-resolved, and build without zeroconf, now example run ok. |
What you have running on your node is not Avahi, but So it is normal that the app outputs:
... because what you have is not Avahi in the first place. I think what you should do is |
Exactly. Until (if ever) we also have an integration with |
Follow-up on #66
The text was updated successfully, but these errors were encountered: