Cloudera Docs

Managing user connections

Create and test per-user credentials for Git repositories and workload APIs so scans and migrations can run without storing passwords in configuration files.

Starting in Cloudera Migration Assistant 4.1.0, the User Connections panel on a cluster's General tab stores credentials for external systems. Each connection belongs to the signed-in user and is scoped to that cluster. Credentials are stored in Vault. Passwords are not returned to the browser after you save a connection.

Supported user connection types depend on cluster type:

  • Workload — control plane API URL and workload user for Cloudera Data Engineering and similar workload sources.

  • Git — repository URL and username for Git-backed migration targets (for example Cloudera Workflow Orchestrator Git workflows).

Agent-managed connection types (Cloudera Manager and control plane endpoints on the agent) are configured separately. You need the Cloudera Migration Assistant administrator role to manage them. You cannot create them from the User Connections panel.

  • The cluster is registered and visible on the Clusters page.

  • Your user has permission to run migrations on that cluster.

  • Vault or the platform credential store is configured for your CMA Master deployment.

  1. Open the cluster and select the General tab.
  2. If no connection exists for the primary type, complete the inline setup form (name, URL, user, and password).
  3. Choose Test Connection. Fix validation errors until the test succeeds.
  4. Choose Register (first-time setup) or + Add Connection (additional Git or Workload connections).

    In the connection drawer, select the connection type, enter fields, test, then choose Register or Update.

  5. Use the row menu to Test, Edit, or Delete connections as needed.

    Status values are Connected, Error, Testing, and Draft, based on the latest test in your session.

When you run scans or migrations that require these credentials, CMA resolves the connection for your user on that cluster. For Cloudera Data Engineering Airflow DAG migration, configure Git connections before mapping DAGs to a target repository. See Migrating Airflow DAGs from to a Cloudera Workflow Orchestrator Git repository.