UPnP: continious errors in logs

[Gooman-rus]

The problem

I have these errors every minute:

2022-06-25 19:59:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:01:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:03:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:05:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:07:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:09:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>
2022-06-25 20:11:04 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=16 read=polling write=<idle, bufsize=0>>

Internet connection is OK. UPnP itself seems to be working fine. So the issue is only in the logs?

What version of Home Assistant Core has the issue?

core-2022.6.7

What was the last working version of Home Assistant Core?

core-2022.5.5

What type of installation are you running?

Home Assistant Supervised

Integration causing the issue

UPnP

Link to integration documentation on our website

https://www.home-assistant.io/integrations/upnp

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

No response

Additional information

No response

[probot-home-assistant]

upnp documentation
upnp source
_{^{(message by IssueLinks)}}

[probot-home-assistant]

Hey there @StevenLooman, @ehendrix23, mind taking a look at this issue as it has been labeled with an integration (upnp) you are listed as a code owner for? Thanks!
_{^{(message by CodeOwnersMention)}}

[ThePatricide]

Same here, although only difference is fd=13 in the logs. It is exactly every 2 minutes that this error pops up. The first time in the logs it shows up there is additional info in the logs though:

2022-06-26 15:07:07 ERROR (MainThread) [async_upnp_client.ssdp] Received error: [Errno 101] Network unreachable, transport: <_SelectorDatagramTransport fd=13 read=polling write=<idle, bufsize=0>>
2022-06-26 15:08:35 WARNING (MainThread) [zeroconf] Error with socket 12 (('fe80::3120:e3b2:4da6:aa34', 5353, 0, 3))): [Errno 101] Network unreachable
Traceback (most recent call last):
File "/usr/local/lib/python3.9/asyncio/selector_events.py", line 1053, in sendto
self._sock.sendto(data, addr)
OSError: [Errno 101] Network unreachable

So it seems for me it's IPv6 related. I think the issue started for me when I disabled IPv6 in the HASS configuration (System -> Network -> IPv6 to disabled). I've re-enabled DHCP (although I don't have IPv6 on my router enabled), I'll let you know what happens.

Edit: seems to have done the trick for me. This error should however not be there when IPv6 is disabled.

[probot-home-assistant]

ssdp documentation
ssdp source
_{^{(message by IssueLinks)}}

[probot-home-assistant]

network documentation
network source
_{^{(message by IssueLinks)}}

[probot-home-assistant]

Hey there @home-assistant/core, mind taking a look at this issue as it has been labeled with an integration (network) you are listed as a code owner for? Thanks!
_{^{(message by CodeOwnersMention)}}

[StevenLooman]

I think when IPv6 is disabled, the adapter(s) found via the network integration shouldn't provide the IPv6 address(es). Not sure what is happening though, perhaps @bdraco has a clue?

[bdraco]

I'm not sure if we actually know if ipv6 is disabled as ifaddr only tells us the addresses not if ipv6 is actually enabled which isn't something we can tell in a cross platform way AFAICT

[StevenLooman]

There is also the question why it reports Network unreachable. Is there a firewall preventing any traffic, for example? Do you have any more information about your set up, @Gooman-rus ?

[balloob]

How are these supervised installations run; bare-metal Ubuntu? VM? k8s?

[ThePatricide]

For me it's bare metal, Intel NUC installation (don't havr an actual NUC though).

[Gooman-rus]

I have Intel NUC with Supervised HA on Ubuntu 20.04. But now I don't have these errors anymore. Firewall is disabled. IPv6 is disabled.

[StevenLooman]

I have Intel NUC with Supervised HA on Ubuntu 20.04. But now I don't have these errors anymore. Firewall is disabled. IPv6 is disabled.

Did you change anything, or just restart home assistant?

[Gooman-rus]

I didn't change anything. I have tried to restart HA and host several times and the errors were still appearing. But after a night I realized that there are no errors anymore.

[mark-miller0]

This issue is happening to me as well. I am running HA 2022.6.7 on macOS 12.4 and python 3.10.5. Previously this was working fine with python 3.8.x and HA 2021.12.x.

I have to restart HA every couple of days to get it working again.

[mark-miller0]

It appears that disabling ssdp is a non-ideal work-around for the issue.

[StevenLooman]

It appears that disabling ssdp is a non-ideal work-around for the issue.

Yes, when you do this SSDP/UPnP devices will no longer be discovered and integrations relying on this will no longer properly work.

[StevenLooman]

@mark-miller0 I'm uncertain if macOS is a supported platform, unless running in a container or VM. Still, I'd like to investigate this. Do you see any changes with regard to the addresses? Windows uses RFC3041 to randomly generate an address and use that, for privacy reasons. Perhaps macOS does this as well, but for link local addresses (i.e., it changes mac address or something)?

Can you see if your IPv6 addresses change between a restart of home assistant, and when the error starts occuring?

[mark-miller0]

@StevenLooman HA has been supported on macOS in a python venv for several years. In my case it has been running fine for 5 years.

It wasn't until the move from python 3.8.9 to 3.10.5 and from HA 2021.12.7 to 2022.6.7 did this happen.

I can take a look at IPv6 this week. macOS is a fairly aggressive user of RFC3041.

[StevenLooman]

@StevenLooman HA has been supported on macOS in a python venv for several years. In my case it has been running fine for 5 years.

You are right, must have read over the link to this page: https://www.home-assistant.io/installation/macos. Though it mentions apt-get, I doubt you'll have that on macOS. 😆 Not sure if macports is still being used, or what else is popular right now, but perhaps somebody with access to macOS could update those instructions.

I can take a look at IPv6 this week. macOS is a fairly aggressive user of RFC3041.

Great, hopefully we can get this fixed.

[mark-miller0]

Yikes! I have not read the current macOS installation instructions for HA in quite a while, but those are much different than what used to be there.

The installation instructions located here are what I have been using with great success:

https://siytek.com/install-home-assistant-on-mac/

Will get back to you soon with some more info on the IPv6 behavior.

[github-actions]

There hasn't been any activity on this issue recently. Due to the high number of incoming GitHub notifications, we have to clean some of the old issues, as many of them have already been resolved with the latest updates.
Please make sure to update to the latest Home Assistant version and check if that solves the issue. Let us know if that works for you by adding a comment 👍
This issue has now been marked as stale and will be closed if no further activity occurs. Thank you for your contributions.