CODEWITHSUNDEEP
pythonubuntuflasksystemdflask-socketio
Nix
Nix

Reputation: 515

Systemd: Start operation timed out. Terminating

I'm trying to create an autostart service for my python-flask-socketio server.

I need to start a python script through systemd. Here's my service code:

[Unit]
Description=AppName


[Service]
Type=forking
ExecStart=/usr/bin/python3 /opt/myapp/app.py

[Install]
WantedBy=multi-user.target

If I try to start it manually using sudo service myservice start - it works just fine. It halts my terminal\ssh window but I can close it and it works like expected.

But when I reboot my PC it does not start. When checking it's status with systemctl status myservice I get the following:

systemd[1]: Starting My Service...
systemd[1]: myserivce.service: Start operation timed out. Terminating.
systemd[1]: Failed to start My Service.
systemd[1]: myserivce.service: Unit entered failed state.
systemd[1]: myserivce.service: Failed with result 'timeout'.

What am I doing wrong here?

Upvotes: 20

Views: 40058

Answers (5)

Brandon Hawbaker
Brandon Hawbaker

Reputation: 338

In my particular case, I also had a flask application and was getting the same error, but it was working originally using the forking type, but recently stopped working after redeploying a new container. In my case, the problem is that the PIDFile property was pointing to a previous container that no longer existed.

In the section: [Service]

There was a property like this causing my issue:

PIDFile=/run/user/1000/containers/overlay-containers/b20a186570e8d177d86a649bab2a81c739ccea9bf5a157b5ea443700871587cf/userdata/conmon.pid

I was able to find the correct container name by running the command:

podman ps

Then I grabbed the first part of the container ID displayed and then matched it to the appropriate subdirectory listed within this location:

/run/user/1000/containers/overlay-containers

After updating it to the correct container ID, I left fork type alone, ran the command:

systemctl --user daemon-reload
systemctl --user start [containername]

Now it functions as expected and no longer times out thinking it's still starting.

Upvotes: 1

Jonas Karl
Jonas Karl

Reputation: 1

It really took me one hour to figure this out :o

Upvotes: -5

Hemingway Hong
Hemingway Hong

Reputation: 7

You have to check /var/log/message.

After you start service, If your script works normally at first and then failed, try this.

ExecStart=/usr/bin/bash -c '/usr/bin/python3 /opt/myapp/app.py'

Upvotes: 0

Voy
Voy

Reputation: 6264

Set a larger start timeout:

[Service]
TimeoutStartSec=300

In case your service would actually need more time to complete the startup.

Upvotes: 14

Anthony Rossi
Anthony Rossi

Reputation: 1262

Your type seems wrong, forking is for programs that detach immediately by themselves. Flask does not, it stays attached to your console.

Your service type should probably be simple

Upvotes: 34

Related Questions