Unstable network connection from inside Home Assistant Docker Container

setfire · February 18, 2026, 10:33am

Hello,

from inside my Home Assistant Docker Container, I experience unreliable network connections to outside of the container. I can reproduce this on two different host machines with different Home Assistant Docker Image Versions.

Other containers on the same host do not show this issue.

Here’s what I do to reproduce the issue:

root@homeserver:~# docker exec -ti homeassistant /bin/bash
home-assistant:/config# nc -vz host.docker.internal 22
nc: host.docker.internal (172.17.0.1:22): Network unreachable
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
nc: host.docker.internal (172.17.0.1:22): Network unreachable
home-assistant:/config# nc -vz host.docker.internal 22
nc: host.docker.internal (172.17.0.1:22): Network unreachable
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
home-assistant:/config# nc -vz host.docker.internal 22
nc: host.docker.internal (172.17.0.1:22): Network unreachable
home-assistant:/config# nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open

From a different container on the same host, I do not see this issue:

root@homeserver:~# docker exec -ti nodered /bin/bash
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open
node-red:~$ nc -vz host.docker.internal 22
host.docker.internal (172.17.0.1:22) open

I also see various failed connections from different integrations in the docker logs for Home Assistant:

2026-02-18 11:21:31.550 ERROR (MainThread) [pyoverkiz.client] Giving up fetch_events(...) after 5 tries (aiohttp.client_exceptions.ClientConnectorError: Cannot connect to host 192.168.1.117:8443 ssl:False [Network unreachable])
2026-02-18 11:21:31.550 ERROR (MainThread) [homeassistant.components.overkiz] Error fetching device events data: Failed to connect.
2026-02-18 11:23:47.759 WARNING (MainThread) [custom_components.device_pulse.coordinator] [Devices XXXXXX] Device [Pixel 8 Pro XXXXXX][192.168.1.133] ping failed but under failure threshold (1/10 failed pings)
2026-02-18 11:24:07.511 WARNING (MainThread) [custom_components.device_pulse.coordinator] [Devices XXXXXX] Device [Pixel 8 Pro XXXXXX][192.168.1.133] ping failed but under failure threshold (1/10 failed pings)
2026-02-18 11:24:39.635 ERROR (MainThread) [homeassistant.components.fritz.coordinator] Error requesting fritz-192.168.1.1-coordinator data: HTTPConnectionPool(host='192.168.1.1', port=49000): Max retries exceeded with url: /upnp/control/hosts (Caused by NewConnectionError("HTTPConnection(host='192.168.1.1', port=49000): Failed to establish a new connection: [Errno 101] Network unreachable"))
2026-02-18 11:26:24.980 ERROR (MainThread) [homeassistant.components.playstation_network.coordinator] Timeout fetching playstation_network data
2026-02-18 11:37:21.246 ERROR (MainThread) [custom_components.daikin_onecta.daikin_api] REQUEST TYPE GET FAILED: Cannot connect to host api.onecta.daikineurope.com:443 ssl:default [Network unreachable]

Here’s the relevant sections of my docker compose file:

name: "container_mgmt"
      
networks:

  smarthome:
    name: "smarthome"
    driver: "bridge"
    labels:
      com.docker.compose.network: "smarthome"
    ipam:
      config:
        - subnet: "172.19.0.0/16"
          gateway: "172.19.0.1"
    
services:

  homeassistant:
    container_name: "homeassistant"
    image: "homeassistant/home-assistant:2026.2.2"
    restart: "unless-stopped"
    hostname: "home-assistant"
    networks:
      smarthome:
        ipv4_address: "172.19.0.7"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    dns:
      - "192.168.1.1"
    ports:
      - "8123:8123/tcp"
    environment:
      - "TZ=Europe/Berlin"
    volumes:
      - "/container/homeassistant/config:/config"
      - "/container/homeassistant/snapshots:/snapshots:rw"
      - "/etc/localtime:/etc/localtime:ro"
    privileged: false
    depends_on:
      - "influxdb"
      - "mosquitto"
    logging:
      options:
        max-size: "5m"
        max-file: "2"

  nodered:
    container_name: "nodered"
    image: "nodered/node-red:4.1.5-22-minimal"
    restart: "unless-stopped"
    hostname: "node-red"
    networks:
      smarthome:
        ipv4_address: "172.19.0.8"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    dns:
      - "192.168.1.1"
    ports:
      - "1880:1880/tcp"
      - "3456:3456/tcp"
    environment:
      - "TZ=Europe/Berlin"
    volumes:
      - "/container/nodered/data:/data"
      - "/container/homeassistant/snapshots:/snapshots:ro"
      - "/container/hostpipe:/hostpipe"
    depends_on:
      - "homeassistant"
      - "mosquitto"
      - "wakeonlan"
    logging:
      options:
        max-size: "5m"
        max-file: "2"

This is my setup:

Host OS: Ubuntu 24.04.4 LTS
Docker Version 28.2.2
Tested Home Assistant Images from 2025.2.2 to 2026.2.2

As I already wrote, I can reproduce this on two differnt host maches. Unfortunately they both have the same OS / Docker installed.

Any help appreciated.

setfire · February 18, 2026, 11:52am

Update: the issue still remains, when changing the Home Assistant Docker containers network mode to “host”.

  homeassistant:
    container_name: "homeassistant"
    image: "homeassistant/home-assistant:2026.2.2"
    restart: "unless-stopped"
    hostname: "home-assistant"
    network_mode: "host"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    dns:
      - "192.168.1.1"
    environment:
      - "TZ=Europe/Berlin"
    volumes:
      - "/container/homeassistant/config:/config"
      - "/container/homeassistant/snapshots:/snapshots:rw"
      - "/etc/localtime:/etc/localtime:ro"
    privileged: false
    depends_on:
      - "influxdb"
      - "mosquitto"
    logging:
      options:
        max-size: "5m"
        max-file: "2"

setfire · February 19, 2026, 5:31pm

It looks like I found the solution myself:
I disabled Offloading and Flow Control on the network adapter of the host machine:

ethtool -K enp1s0 tso off gso off gro off lro off rx off tx off
ethtool -A enp1s0 rx off tx off

No more “Network unreachable”.

fkuiper · February 20, 2026, 7:59am

I’ve had the same problem yesterday. Disabling offloading and flow control did not fix it however. I ended up rolling back the latest kernel update and now my home assistant runs stable again.

I’m back running Ubuntu 24.04.4 LTS (GNU/Linux 6.8.0-88-generic x86_64) again. Look like 6.8.0-100 has some issues on my mini pc.

setfire · February 20, 2026, 7:04pm

Well, I guess you’re right. Changing the interface’s parameters only fixed it temporarily for me. Maybe we should file a bug report. But I don’t know where.

I rolled back to 6.8.0-90-generic and that fixed the issue as well.

maks · February 21, 2026, 8:45am

I had exactly the same issue. Also disabled hw offloading but it did not work. Now that you are mentioning mini pc, I am running this on an age old Zotac mini pc (CI323) with ancient Realtek NICs.

Als curious to understand why this is only a problem in HA. All other containers are running fine.

setfire · February 21, 2026, 3:35pm

In have a MeLE Quieter HD3 Mini PC. You’re right, it’s really interesting that only the Home Assistant Container has the issue. But I would see it more as an Ubuntu or Kernel issue since up to the last Kernel patch everything was fine.

M66B · February 21, 2026, 4:41pm

This might be the same issue:

github.com/home-assistant/core

Ubuntu 24.04 Kernel 6.8.0-100 breaks UDP multiple integrations with PermissionError

opened 12:11PM - 09 Feb 26 UTC

fanderko

### The problem I encountered issues with multiple integrations (GoodWe, Tuya, …Ariston, HACS). Any integration attempting to use UDP or network discovery fails with PermissionError or Network Unreachable. This is a system-wide networking issue affecting the Docker host mode over Wi-Fi. ### What version of Home Assistant Core has the issue? core-2026.2.1 ### What was the last working version of Home Assistant Core? core-2026.2.1 ### What type of installation are you running? Home Assistant Container ### Integration causing the issue GoodWe, Ariston, HACS, Tuya ### Link to integration documentation on our website _No response_ ### Diagnostics information _No response_ ### Example YAML snippet ```yaml ``` ### Anything in the logs that might be useful for us? ```txt Logger: homeassistant.config_entries Source: config_entries.py:749 First occurred: 9:58:37 PM (1 occurrence) Last logged: 9:58:37 PM Error setting up entry GoodWe for goodweTraceback (most recent call last): File "/usr/src/homeassistant/homeassistant/config_entries.py", line 749, in __async_setup_with_context result = await component.async_setup_entry(hass, self) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/src/homeassistant/homeassistant/components/goodwe/__init__.py", line 21, in async_setup_entry inverter = await connect( ^^^^^^^^^^^^^^ ...<3 lines>... ) ^ File "/usr/local/lib/python3.13/site-packages/goodwe/__init__.py", line 56, in connect await inv.read_device_info() File "/usr/local/lib/python3.13/site-packages/goodwe/et.py", line 461, in read_device_info response = await self._read_from_socket(self._READ_DEVICE_VERSION_INFO) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/goodwe/inverter.py", line 132, in _read_from_socket result = await command.execute(self.host, self.timeout, self.retries) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/goodwe/protocol.py", line 149, in execute transport, _ = await loop.create_datagram_endpoint( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<2 lines>... ) ^ File "/usr/local/lib/python3.13/asyncio/base_events.py", line 1461, in create_datagram_endpoint raise exceptions[0] File "/usr/local/lib/python3.13/asyncio/base_events.py", line 1448, in create_datagram_endpoint await self.sock_connect(sock, remote_address) File "/usr/local/lib/python3.13/asyncio/selector_events.py", line 641, in sock_connect return await fut ^^^^^^^^^ File "/usr/local/lib/python3.13/asyncio/selector_events.py", line 649, in _sock_connect sock.connect(address) ~~~~~~~~~~~~^^^^^^^^^ PermissionError: [Errno 13] Permission denied Logger: custom_components.ariston Source: custom_components/ariston/__init__.py:88 integration: Ariston (documentation, issues) First occurred: 9:58:34 PM (2 occurrences) Last logged: 9:58:55 PM Traceback (most recent call last): File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 1248, in _wrap_create_connection sock = await aiohappyeyeballs.start_connection( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<6 lines>... ) ^ File "/usr/local/lib/python3.13/site-packages/aiohappyeyeballs/impl.py", line 122, in start_connection raise first_exception File "/usr/local/lib/python3.13/site-packages/aiohappyeyeballs/impl.py", line 73, in start_connection sock = await _connect_sock( ^^^^^^^^^^^^^^^^^^^^ ...<6 lines>... ) ^ File "/usr/local/lib/python3.13/site-packages/aiohappyeyeballs/impl.py", line 208, in _connect_sock await loop.sock_connect(sock, address) File "/usr/local/lib/python3.13/asyncio/selector_events.py", line 641, in sock_connect return await fut ^^^^^^^^^ File "/usr/local/lib/python3.13/asyncio/selector_events.py", line 649, in _sock_connect sock.connect(address) ~~~~~~~~~~~~^^^^^^^^^ OSError: [Errno 101] Network unreachable The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/config/custom_components/ariston/__init__.py", line 88, in async_setup_entry device = await ariston.async_hello( ^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<2 lines>... ) ^ File "/usr/local/lib/python3.13/site-packages/ariston/__init__.py", line 78, in async_hello await self.async_discover() File "/usr/local/lib/python3.13/site-packages/ariston/__init__.py", line 65, in async_discover cloud_devices = await _async_discover(self.api) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/ariston/__init__.py", line 187, in _async_discover ] = await asyncio.gather( ^^^^^^^^^^^^^^^^^^^^^ api.async_get_detailed_devices(), api.async_get_detailed_velis_devices() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/usr/local/lib/python3.13/site-packages/ariston/ariston_api.py", line 551, in async_get_detailed_velis_devices detailed_velis_devices = await self._async_get( ^^^^^^^^^^^^^^^^^^^^^^ f"{self.__api_url}{ARISTON_VELIS}/{ARISTON_PLANTS}" ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/usr/local/lib/python3.13/site-packages/ariston/ariston_api.py", line 992, in _async_get return await self.__async_request("GET", path, params, None) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/ariston/ariston_api.py", line 948, in __async_request response = await session.request( ^^^^^^^^^^^^^^^^^^^^^^ method, path, params=params, json=body, headers=headers ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/usr/local/lib/python3.13/site-packages/aiohttp/client.py", line 770, in _request resp = await handler(req) ^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/aiohttp/client.py", line 725, in _connect_and_send_request conn = await self._connector.connect( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ req, traces=traces, timeout=real_timeout ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 622, in connect proto = await self._create_connection(req, traces, timeout) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 1189, in _create_connection _, proto = await self._create_direct_connection(req, traces, timeout) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 1561, in _create_direct_connection raise last_exc File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 1530, in _create_direct_connection transp, proto = await self._wrap_create_connection( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<7 lines>... ) ^ File "/usr/local/lib/python3.13/site-packages/aiohttp/connector.py", line 1271, in _wrap_create_connection raise client_error(req.connection_key, exc) from exc aiohttp.client_exceptions.ClientConnectorError: Cannot connect to host www.ariston-net.remotethermo.com:443 ssl:default [Network unreachable] ``` ### Additional information This is a host-level kernel regression. If you encounter this, downgrading the kernel to 6.8.0-94-generic restores full functionality. This seems to be related to how the new kernel handles unprivileged UDP sockets over Wi-Fi.

maks · February 22, 2026, 2:02pm

That makes a lot of sense, I also tried a USB network adapter and experienced the same issues. Hopefully the 103 kernel will fix it.

Edit\ Updated to 6.8.0-103-generic and this fixed the issue, not GA yet though

setfire · February 22, 2026, 2:55pm

Well, that’s good news. Seems like enough people were complaining
I will try the update next week because I’m on vacation.

Normann_P_Nielsen · March 2, 2026, 7:41pm

Jsut wanna mention I have the same issues - only from the HA container.

Also on 6.8.0-100-generic #100-Ubuntu SMP PREEMPT_DYNAMIC

gadd · March 5, 2026, 12:38pm

wasted several days trying to find a solution… that drove me crazy. Glad I’ve found this post. rolled back to 6.8.0-88-generic