latest OS 2021.8.5 and got this single error which I’m not sure where it coming from.
Error with socket 90 (('fe80::a6a1:b841:7055:dea3', 5353, 0, 2))): [Errno 101] Network unreachable
Any idea?
Port 5353 is mDNS. Are all your containers running ?
so far all my container is up and running… nothing is stop working at the moment. Thats is why I get confused.
I am also getting this error since the last update. The socket changes with each reboot.
Running Operating system in Proxmox VM. Other containers all running (unifi controller; pihole; OMV).
Everything appears to be working - although restarts are taking twice as long.
Source: /usr/local/lib/python3.9/site-packages/zeroconf/_logger.py:74
First occurred: 12:27:39 (1 occurrences)
Last logged: 12:27:39
Error with socket 27 (('fe80::fc20:7167:57f0:855', 5353, 0, 3))): [Errno 101] Network unreachable
Traceback (most recent call last):
File "/usr/local/lib/python3.9/asyncio/selector_events.py", line 1056, in sendto
self._sock.sendto(data, addr)
OSError: [Errno 101] Network unreachable
Anyone know where I can look to find out what is causing?
Thanks.
so far one thing in common… we both running proxmox
Roll back to 2021.8.4 fixes the issue.
Now to find out what is causing this…
Nothing is jumping out from the 2021.8.5 breaking changes and my current setup:
- only slight hint is that zeroconf version was bumped (and my error contains a reference to zeroconf).
Going to 2021.8.6 with no changes… the error continue
furthermore coronavirus integrations fail to get curent and recovered info.
Slight update:
I reinstalled 2012.8.5 and error came back.
Believe the port referenced in the issue (5353) is used by apple bonjour and chromecast devices.
If I comment out zerconf: from my configuration.yaml file then the issue disappears - mDNS is enabled on my USG.
Now to find out what I am going to miss by not having zerconf in my configuration, anyone able to enlighten me? or alternatively what I need to fix to remedy the problem.
I think it probably has something to do with my firewall rules for my subnets (as my chromecasts are on a different subnet to HA, but a rule is in place to allow to access port 5353).
I dont have zerconf and it happen to me bro.Also not running firewall on different subnets either
1 Like
Zeroconf is included in Default Config. Or you have removed default_config and also remove zeroconf?
Have always operated with Default Config removed from my configuration file - zeroconf was added back in manually.
Then today when attempting to track down the issue I commented out zeroconf.
(I also added default config: back into configuration file and the error still occurs)
It doesn’t only happen on Proxmox installs.
2012.8.5 & 2012.8.6 Supervised on RPi4 with Debian 10:
Logger: zeroconf
Source: /usr/local/lib/python3.9/site-packages/zeroconf/_logger.py:74
First occurred: 17:14:07 (1 occurrences)
Last logged: 17:14:07
Error with socket 57 (('fe80::5ba:f4eb:c053:4448', 5353, 0, 2))): [Errno 101] Network unreachable
Traceback (most recent call last):
File "/usr/local/lib/python3.9/asyncio/selector_events.py", line 1056, in sendto
self._sock.sendto(data, addr)
OSError: [Errno 101] Network unreachable
What caught my attention:
Error with socket 57 (('fe80::5ba:f4eb:c053:4448', 5353, 0, 2))): [Errno 101] Network unreachable
Not running any IPv6-network here, all purely on IPv4 only and the LAN is configured accordingly. IPv6 is disabled on all interfaces.
According to Zero-configuration networking:
“IPv6 will automatically be enabled if one of the selected interfaces has an IPv6 address that is enabled via the Network integration.”
How to disable IPv6 searches for Zeroconf?
This happend to me also - home assistant running on proxmox. On prev. releases I didnt noticed such error. Didnt tuch config about zeroconf. IPv4 is set to static and IPv6 dynamic.
An issue has been opened on github:
1 Like
I lost discovery of added devices/ integrations. Ie) cannot add in new ESPhome devices.
So I have added zeroconf back as the error message is less of a problems than losing discovery of devices.
For the list: AlexxIT/SonoffLAN is also not working in local mode if Zeroconf is disabled (Issue).
1 Like
For zeroconf, here is a potential fix if anyone can give it a try.
1 Like
Happy to try it out - only problem is - I don’t know how to do that?
Do I edit the 3 individual files on my system with the proposed changes? or are there files that I download and copy across?
Tried to install the 0.35.0 version but it didn’t solve the problem
Zeroconf 0.35.0 fixes the problem for me.
Running 2021.9.0.dev20210815