Resolving sandbox restarts caused by nc not found

A sandbox can restart repeatedly when its entrypoint.sh script cannot start the sandbox API. The usual causes are a missing netcat package or failed localhost resolution in a Debian-based image.

Remove the entrypoint script

If you do not need custom initialization, remove entrypoint.sh and start the sandbox API directly:

ENTRYPOINT ["/usr/local/bin/sandbox-api"]

Remove the custom entrypoint instructions:

COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh
ENTRYPOINT ["/entrypoint.sh"]

Fix the entrypoint script

If you need custom initialization, install netcat-openbsd in the Dockerfile:

RUN apt-get update \
    && apt-get install --assume-yes --no-install-recommends --quiet \
        netcat-openbsd \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

Then replace localhost with 127.0.0.1 in the port check:

while ! nc -z 127.0.0.1 "$port"; do
  sleep 1
done

Debian-based images do not always resolve localhost automatically. Use 127.0.0.1 for local connections in these images.

Inspect sandbox logs

Check the console logs for errors from the entrypoint script:

bl logs sbx your-sandbox-name

Sandbox processes

Learn how to start and manage sandbox processes.

Docker sandbox tutorial

Learn how to configure Docker workloads in sandboxes.

Get Started

Sandboxes

Agent Drive

Volumes

Batch Jobs

Applications

MCP Servers Hosting

Model Gateway

Observability

Integrations

Administration & security

Troubleshooting

Resolving sandbox restarts caused by nc not found

Remove the entrypoint script

Fix the entrypoint script

Inspect sandbox logs

Sandbox processes

Docker sandbox tutorial

​Remove the entrypoint script

​Fix the entrypoint script

​Inspect sandbox logs

Sandbox processes

Docker sandbox tutorial

Remove the entrypoint script

Fix the entrypoint script

Inspect sandbox logs