Core Concepts

Understanding these concepts will help you use jokowipe.id effectively.

Backup

A backup is a point-in-time snapshot of a database. jokowipe.id creates backups by running native database tools (pg_dump, mysqldump, mongodump, etc.) on the agent machine and uploading the result to your configured storage target.

Each backup record contains:

• Backup ID — Unique identifier (e.g., bkp-a1b2c3d4)
• Status — pending, running, success, failed, expired
• Database URL — The source database (credentials are never stored on our servers)
• Storage path — Where the backup file lives in your storage target
• Size — Compressed size of the backup file
• Duration — How long the backup took
• Checksum — SHA-256 hash for integrity verification
• Metadata — Tags, retention policy, schedule ID (if triggered by a schedule)

Schedule

A schedule defines when automatic backups run. jokowipe.id supports:

Type	Example	Description
Cron	0 2 * * *	Run at 2:00 AM every day
Interval	every 6h	Run every 6 hours
One-shot	(via API)	Run once at a specific time

Schedules are associated with:

• An agent (where to run the backup)
• A database (what to back up)
• A storage target (where to store it)
• A retention policy (how long to keep backups)

Storage Target

A storage target is a cloud storage bucket or local path where backup files are stored. jokowipe.id supports:

• AWS S3
• Google Cloud Storage (GCS)
• Cloudflare R2
• Azure Blob Storage
• MinIO (self-hosted S3-compatible)
• Local filesystem (agent-local, not recommended for production)

Bring Your Own Storage

jokowipe.id uses a BYOS (Bring Your Own Storage) model. Your backup files are uploaded directly from the agent to your storage bucket. The jokowipe.id API never receives or stores your backup data — only metadata (backup size, duration, checksum, storage path).

Agent

The agent is a lightweight binary that runs on your database server or a machine with database access. It:

1. Polls the jokowipe.id API for scheduled backup jobs
2. Executes the backup command locally
3. Uploads the backup file to your storage target
4. Reports status and metadata back to the API

Agents authenticate using agent tokens — short-lived JWTs signed by the API. They communicate outbound-only over HTTPS, so no inbound firewall rules are needed.

Policy

A retention policy defines how long backup files are kept. You can set:

• Count-based: Keep the last N backups
• Age-based: Keep backups for N days
• Combined: Keep at least N backups and at most N days

When a backup expires, the agent (on its next run) or the API (via a scheduled cleanup job) deletes the file from your storage target.

Organization

An organization is the top-level entity in jokowipe.id. All agents, storage targets, schedules, and backups belong to an organization.

Team

Teams are groups of users within an organization with specific roles:

Role	Permissions
Owner	Full access, billing, delete org
Admin	Create/delete all resources, manage members
Member	Create/manage backups and schedules
Viewer	Read-only access to backup history

Anomaly

An anomaly is an unusual backup pattern detected by the AI system. Examples:

• Backup size suddenly drops by >50% (data loss risk)
• Backup duration doubles (performance degradation)
• Backup failed multiple times in a row
• No backup completed within the expected window

Anomalies trigger alerts via the configured notification channels.

Restore

A restore operation downloads a backup file and applies it to a target database. jokowipe.id supports:

• Download restore — Agent downloads and restores the backup to a local database
• Remote restore — Restore to a different database instance
• Point-in-time restore — Browse backup history and restore to any snapshot

Webhook

A webhook is an HTTP callback that jokowipe.id calls when specific events occur (backup success, failure, anomaly detected, etc.). You can use webhooks to integrate with external systems.