Docker: Restarting loop

[jrieger]

The problem

New installation of docker container on ARM results in restarting loop:

INFO: Home Assistant Core finish process exit code 256
INFO: Home Assistant Core finish process received signal 11
…

What version of Home Assistant Core has the issue?

2022.7.1

What was the last working version of Home Assistant Core?

2022.6.7

What type of installation are you running?

Home Assistant Container

Integration causing the issue

No response

Link to integration documentation on our website

No response

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

No response

Additional information

Running on QNAP NAS (TS-431XeU).

Processor: Annapurna Labs Alpine AL314 Quad-core ARM Cortex-A15 CPU @ 1.70GH

[frenck]

Please provide a reproduction scenario, e.g., providing docker CLI used, Docker composes file.
There should be more log too, could you provide those?

[jrieger]

Docker compose file:

version: '3'
services:
  homeassistant:
    container_name: homeassistant-test
    hostname: homeassistant-test
    image: homeassistant/home-assistant:stable
    restart: unless-stopped
    privileged: true
    networks:
      backend:
        ipv4_address: 10.10.0.99
    environment:
      DISABLE_JEMALLOC: true
      TZ: Europe/Berlin
networks:
  backend:
    name: backend
    driver: bridge
    ipam:
      driver: default
      config:
        - subnet: 10.10.0.0/24
          gateway: 10.10.0.1

No logs, /config is empty.

Executing python3 -m homeassistant --config /config inside container results in Segmentation fault.

[frenck]

Our first thought is that this might be related to your system running/having a really old Linux kernel (acient level, also driven by the fact that jemalloc has been disabled, which is a good indicator for that).

Which kernel version does your system run?

[jrieger]

Linux det 4.2.8 #2 SMP Tue May 31 05:27:29 CST 2022 armv7l unknown

2022.6.7 is running fine, though.

[frenck]

So, that is pretty much acient history (2015) and is what causes this issue. From the Home Assistant perspective, there is not much we can do about that.

Upgrading the kernel of your system is the only route to go.

[jrieger]

Kindly understand that, since this is a NAS system, I have no control on the kernel version. So this is the list of QNAP devices that stop supporting Home Assistant. Can you point me to the change that caused this issue?

[frenck]

There isn't directly a single change that can be appointed to this, but the most likely one is that we upgraded our base images from Alpine Linux 3.14 to 3.16 (since 3.14 is EOL).

[jrieger]

OK, thanks for your effort and information. I'll try to contact QNAP, or migrate to another system anyways.

[frenck]

👍 Sorry that I haven't got better news for you

[brozikcz]

I suppose I have the same issue on my Odroid C2 with 3.16.85-65 kernel when I'm getting:

[21:31:51] INFO: Home Assistant Core finish process exit code 256
[21:31:51] INFO: Home Assistant Core finish process received signal 4

without any errors.

[larsxschneider]

@jrieger @brozikcz I ran into the exact same issue on my QNAP device. Obviously I also have the same old Linux version on the QNAP.

Where you able to find a workaround or where you able to debug further and pin point the issue?

[Wetzel402]

I also have this issue running kernel version 4.2.8 on a QNAP TS-431XeU running the latest QTS 5.0.1.2079 firmware.

[larsxschneider]

@Wetzel402 I have the same machine 😄

I think the core problem might be a bad piwheels package that creeped into the HA build:
#75142 (comment)
piwheels/packages#276 (comment)

If that is indeed the problem, then the situation should fix itself once HA updates piwheel again.

You could also try to build wheels yourself for your HA (I haven't done that yet because I had not enough time yet):
piwheels/packages#276 (comment)

[Wetzel402]

@larsxschneider,

Nice.

I hope that it is something simple like that. I would hate to see newer HA releases not working on my NAS. With that said, it sounds like this is a ARMv6 issue but isn't the Alpine AL 314 in our NAS ARMv7?

[lswysocki]

Hello ALL,
I have TS-431XeU and the same problem, as I understand the only solution is to install the older version 2022.6.7?

[lswysocki]

Image: homeassistant/home-assistant:2022.6.7
Console: s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service legacy-services: starting
services-up: info: copying legacy longrun home-assistant (no readiness notification)
s6-rc: info: service legacy-services successfully started
: Unsupported system page size
: Unsupported system page size
: Unsupported system page size
Fatal Python error: _PyRuntimeState_Init_impl: Can't initialize threads for interpreter
Python runtime state: unknown

[09:18:36] INFO: Home Assistant Core finish process exit code 1
[09:18:36] INFO: Home Assistant Core service shutdown
s6-rc: info: service legacy-services: stopping
s6-rc: info: service legacy-services successfully stopped
s6-rc: info: service legacy-cont-init: stopping
s6-rc: info: service legacy-cont-init successfully stopped
s6-rc: info: service fix-attrs: stopping
s6-rc: info: service fix-attrs successfully stopped
s6-rc: info: service s6rc-oneshot-runner: stopping
s6-rc: info: service s6rc-oneshot-runner successfully stopped

[larsxschneider]

@lswysocki I am currently running the docker image homeassistant/armhf-homeassistant:2022.6.7 on my TS-431XeU without issues. I haven't done any more debugging on the newer versions though.

[lswysocki]

@larsxschneider,
Unfortunately the HA container does not start.

[larsxschneider]

I am running firmware version 5.0.0.1986 and Container Station version V2.6.0.483. How about you?

[lswysocki]

QNAP: 5.0.0.2055
Container Station: 2.6.1.544

[Wetzel402]

This is my docker compose at the moment...

services:
  homeassistant:
    container_name: homeassistant
    image: "ghcr.io/home-assistant/home-assistant:2022.6"
    volumes:
      - /share/Container/home-assistant:/config
    restart: unless-stopped
    privileged: true
    ports:
      - 8123:8123
    environment:
        DISABLE_JEMALLOC: true
        TZ: America/Chicago
    networks:
      backend:
        ipv4_address: 10.10.0.99
networks:
  backend:
    name: backend
    driver: bridge
    ipam:
      driver: default
      config:
        - subnet: 10.10.0.0/24
          gateway: 10.10.0.1

Version 2022.7+ currently boot loop with no known fix.

[Wetzel402]

I can confirm the issue exists in version 2022.9 also.