Troubleshoot Pipelab

Use the recovery steps that match the screen or command that failed. If an issue comes from a provider, include its name and the relevant log excerpt when you ask for help. Remove credentials and private paths first.

Desktop startup

The desktop app starts or connects to a local CLI server before the dashboard loads. If the app stays on Connecting or shows Disconnected:

1. Use the retry action on the screen.
2. If you started the UI for development, confirm both the UI dev server and CLI backend are running; the desktop package starts these for you.
3. If startup data fails to load, retry after the connection is restored.
4. If packaged desktop startup reports a CLI/server failure, restart the app and collect the displayed error for diagnosis.

Pipeline validation and execution

• Required parameter is missing: Open the task editor and complete the required fields shown by that task.
• Plugin or node is unavailable: The bundled desktop task picker does not install or activate plugins. Choose an action available in this build or remove the unavailable task from the graph.
• Advanced value becomes empty: Pipeline parameter evaluation logs errors and substitutes an empty string for that parameter. Review the task log and expression inputs.
• A later task did not run: Legacy graph execution stops when an action fails. Fix the failing task before running the pipeline again.

Release workflow planning

• Needs attention: Open the issue detail on the source, build, destination, or deployment slot that reports it. Check its required provider fields and selected artifact compatibility.
• Build target is unavailable: Availability depends on the selected producer and current host platform/architecture. Use only targets listed by that producer.
• Dry-run fails: pipelab workflow run <id-or-name> --dry-run still loads providers and validates the saved workflow. A dry-run does not execute deployment tasks, but it can report missing settings or invalid references.

CLI and provider failures

• pipelab run requires a readable JSON file with graph or canvas.blocks and any referenced plugins available to the CLI.
• pipelab workflow run requires a saved workflow. Use an ID or a unique name; use --dry-run to inspect planning issues before execution.
• Provider errors often indicate missing account connections, external tools, project settings, or host limitations. Follow the provider page for the current requirements.

See the CLI reference, provider catalog, and Runs and history for command and execution details.