Skip to content

Deploys, statuses and logs

Deploy

Deploy creates a new deployment: BotBarn fetches your code, builds an image and starts a container. The previous deployment of the bot is stopped. Deploy again whenever you change the code, the source or the variables.

The Deployments tab lists every deployment, newest first.

Statuses

Status Meaning
Queued Waiting for the runner to pick it up, usually a few seconds.
Building Fetching the code and building the image.
Starting The container started. BotBarn watches it for 15 seconds.
Running The bot has been up for 15 seconds.
Restarting The bot exited after it was running; BotBarn is starting it again.
Stopping / Stopped You stopped it, its plan ended, or a newer deployment replaced it.
Failed The build failed, the bot exited during startup, or it crashed too many times. The reason is on the Overview and in Logs.

Crashes and automatic restarts

  • A bot that exits within 15 seconds of starting is marked Failed with Bot exited during startup (exit code N). It is not restarted: something is wrong with the code or its settings. Read the log.
  • A bot that exits later is restarted automatically, up to 5 times. The sixth crash marks it Failed with Bot crashed 6 times.
  • Once the bot reaches Running after a restart, the count goes back to 0.

Stop, start and restart

Action What happens
Stop Stops the container. The bot stays off until you start it.
Start Starts a stopped deployment again: same code, current variables, restart count reset.
Restart Restarts the running container: same code, same variables, same /data.
Deploy New code, new image, current variables.

Start, Restart and Deploy need an active plan. Stop always works.

Logs

The Logs tab shows:

  • the build log of the deployment: the generated or used Containerfile and the build output;
  • the bot log: the last 200 lines your bot printed to standard output and standard error, refreshed while it runs.

Values of your variables are shown as ***.

Tips:

  • Print to standard output. In Python, BotBarn runs your code unbuffered; with your own Containerfile set PYTHONUNBUFFERED=1 or log with flush=True.
  • Log one line per event, with the error message. A crash's last lines are the most useful ones.

Data

Files your bot writes in /data are kept across restarts and deploys; files written anywhere else (/tmp, /var/tmp) are lost when the container restarts. /data is sized by the plan and is deleted with the bot and 7 days after the plan ends. Download or reset it from the Data card of the bot's Overview or with your AI assistant. A reset stops the bot, empties /data and starts it again.