Operations

Operations

Edit page
Run, monitor, recover, and troubleshoot a production Customer Portal deployment.

These runbooks cover the work after Customer Portal has been deployed: protecting its data, detecting failures, diagnosing common problems, and recovering safely from an unsuccessful upgrade.

Backup and restore

Define recovery targets, protect a PostgreSQL backup, and prove that it can be restored.

Troubleshooting

Diagnose startup, database, authentication, email, and layer-discovery failures.

Observability

Understand the signals available today and the minimum production monitoring baseline.

Upgrade recovery

Respond to failed migrations, application regressions, and schema compatibility problems.

Own the recovery contract

Every deployment should name an operator, define a recovery point objective (how much data may be lost), and define a recovery time objective (how long recovery may take). The deployment owner also needs access to the source revision, database backups, platform secrets, DNS, and email-provider configuration before an incident begins.

These pages describe the current repository behavior. Adapt the commands to the database and hosting platform, then rehearse them against non-production infrastructure.