Deploy and operate LeapView

LeapView separates application delivery from project delivery. Choose the workflow that matches what is changing, then use the focused operational guide for execution and verification.

Deliver the application

Application delivery changes the LeapView binary or image, browser assets, runtime configuration, infrastructure, and persistent storage attachment.

• Use Production configuration to establish secrets, public addresses, storage, and capacity boundaries.
• Use Self-hosting to deploy the supported single-node topology.
• Use Upgrades and migrations to coordinate provider-managed image changes, persistent migrations, and retained-generation rollback.

Application releases should use immutable artifacts and should not silently modify project resources.

Deliver a project

Project delivery changes the source-root connections, sources, Models, semantic models, pipelines, dashboards, and managed-data revision pins; target-owned access policy and publication state remain separate controls.

• Use Develop, review, and publish for the exact-candidate delivery workflow.
• Use Targets and environments to keep local, staging, and production identities explicit.
• Read Projects and environments for the ownership and activation model behind that workflow.

Promote the same reviewed source-root commit and managed revision identities through environments rather than maintaining separate dashboard trees.

Run and recover the service

• Use Health and observability for readiness, metrics, logs, refresh activity, query events, and synthetic verification.
• Use PostgreSQL operations and high availability for pool alerts, provider telemetry, maintenance fencing, and failover validation.
• Use Backup and restore to coordinate provider-native recovery points; LeapView does not create a production backup archive.
• Use Delivery reachability and recovery boundaries when an immutable delivery root needs investigation or recovery.
• Use Operational troubleshooting to locate failures across infrastructure, active data, semantic modeling, and dashboard behavior.
• Use Storage and recovery for local and object-storage analytical boundaries.

For accepted settings and flags, use the generated Environment variable reference and CLI command reference.