Frequently asked questions

Short answers with a link to the detail.

I don’t have VPN, can I work?

Yes. Resolve as mock the dependencies on cores you can’t reach. You only need a connection if you leave some dependency as cloud. See Mock a dependency.

Do I need Docker?

To launch the infrastructure (Kafka/Redis/Postgres) or run micros in container mode, yes. You can work in native mode without it if your scenario doesn’t declare infra. See Infrastructure.

All my micros declare port 8080, do they collide?

No. MicroLab remaps the ports at runtime to free effective ports. See Effective ports and URL rewriting.

Does MicroLab modify my repos?

Never. All the configuration (URLs, properties, database) is applied by injection at startup. The non-versioned local part lives in settings.local.yml.

A micro won’t start and I don’t see anything in its logs

When a micro doesn’t manage to start, the reason is in the engine logs, not its own. Use the diagnosis (microlab diagnose <micro>) and filter the logs by engine. See Debug a micro that won’t start.

How do I point a micro at another database?

With the database selector in its editor (local, cloud, custom or a connection from the catalog). See Point at another DB.

I switched branches and something doesn’t add up

Update the manifest from the repository: it reconciles the drift keeping your customization. See Switch branches and drift.

Can I pass a scenario to a teammate?

Yes, by exporting it to a self-contained file; on importing, it reconciles without overwriting their catalog. See Share a scenario.

Can I do all of this from the terminal or with an AI?

Yes. The microlab CLI does the same as the app, and the MCP server and the copilot expose it to AI agents.

What data does MicroLab send?

Only diagnostics of its own errors, anonymous and without your logs or secrets. It’s opt-out. See Privacy and telemetry.