Settings page produces a 504

1

I’ve just upgraded to the docker containers on a raspberry pi 3 to the “latest”. The settings page now says master@19dd33d.

And since the update the web frontend is broken. The schedule overview and the system info page work fine. But when I click on settings I get a 504 timeout after a while.

What could be broken here?

2

@SteffenK, I’m looking into it at the moment. For the time being, I’d like to know about your environment setup:

  • Raspberry Pi version
  • OS (whether it’s Raspberry Pi OS Lite or Balena)
  • How did you upgrade Anthias?

Would you also be able to provide server logs? Details on how to do it can be found here.

Thanks!

3
  • It’s a Raspberry Pi 3 Model B Plus Rev 1.3
  • The OS is Bullseye and it’s an install that was done with a Lite image and the installation script.
  • Using the upgrade_containers.sh script

As for the logs: When the 504 happens in the webbrowser the container screenly-anthias-server-1 says:

host_agent_service is not ready after 60 retries

And I don’t know if this is somehow related but in the Celery container I can see:

[2024-07-17 07:50:42,550: INFO/Beat] Scheduler: Sending due task display_power (celery_tasks.get_display_power)
[2024-07-17 07:50:42,615: INFO/MainProcess] Task celery_tasks.get_display_power[c5f484d4-7ee5-4f11-b2f6-1f1bdfa6d423] received
[2024-07-17 07:51:12,662: ERROR/MainProcess] Task handler raised error: TimeLimitExceeded(30)
Traceback (most recent call last):
  File "/usr/local/lib/python3.7/dist-packages/billiard/pool.py", line 684, in on_hard_timeout
    raise TimeLimitExceeded(job._timeout)
billiard.exceptions.TimeLimitExceeded: TimeLimitExceeded(30,)
[2024-07-17 07:51:12,665: ERROR/MainProcess] Hard time limit (30s) exceeded for celery_tasks.get_display_power[c5f484d4-7ee5-4f11-b2f6-1f1bdfa6d423]
[2024-07-17 07:51:14,376: ERROR/MainProcess] Process 'ForkPoolWorker-2' pid:22 exited with 'signal 9 (SIGKILL)'
4

Broken settings here also, fresh install.

I don’t know if this has anything to do with it, but journalctl reports.
Started Anthias Host Agent.
Traceback (most recent call last):
File “/home/root/screenly/host_agent.py”, line 10, in
import netifaces
ModuleNotFoundError: No module named ‘netifaces’
anthias-host-agent.service: Main process exited, code=exited, status=1/FA>
anthias-host-agent.service: Failed with result ‘exit-code’.

5

@Dhe, got it, and will take a look at it. Is that on Bullseye?

6

@SteffenK, in theory those failed retries shouldn’t get in the way of the settings page, but I’ll take a closer look at it.

7

image

8

Since, you’re doing a fresh install, you can try Bookworm at the moment while I investigate the issue.

9

@Dhe, @SteffenK fixes are underway — see Anthias PR #1979 for details. I’m currently testing my changes. I’ll let you know when the fix is already shipped.

The pull request was already merged. You can now test the install script.
(Make sure to back up your data/assets first before doing so.)

10

Works great!
Thanks for the fast fix.

1 Like
11

I’ve also upgraded mine and it works again.

Thanks!

1 Like