Backup & Recovery Guide

Backup & Recovery protocols are mission-critical for protecting against data loss, disaster, accidental deletion, and for compliance with retention and business continuity policies. This page outlines how to back up and restore InnoSynth-Forjinn, including types of data, scheduling, best practices, and restore drills.

---

What to Back Up

• Database (DB): All workflow configs, agent setups, user/workspace data, logs, and metadata (typically Postgres, MySQL, etc).
• Uploads/Artifacts: All user-uploaded files, document blobs, images, and any stored datasets.
• System Config & Secrets: .env files, K8s Secrets/ConfigMaps, deployment yamls, credential stores (if file-based).
• Logs: Optionally, keep access/error/app logs for compliance or incident review.

---

Creating Backups

Database

• Use DB-native tools (pg_dump, mysqldump, managed cloud backup) to export data daily/hourly.
• For cloud-managed DB: Enable automated backups with a 7–30 day retention window.
• Store dumps in a secure, offsite bucket (S3, Azure Blob, GCS, or NFS/remote mount).

Files/Uploads

• Regularly (e.g., hourly or nightly) rsync or copy the /uploads or /data directory to a backup target.
• For cloud, use Storage Lifecycle/Archival policies.

System Config

• Encrypt and back up deployment/environment secrets in a secure key vault or secret store.
• Export Helm releases or Docker Compose YAMLs.

---

Scheduling/Automation

• Recommended: Use cronjobs, CI workflows, or cloud-native schedulers.
• Test regular restores automatically in staging to validate recoverability (restore drill).

---

Restore Procedure

1. Restore DB: Stop all services; restore DB dump into a clean DB instance.
2. Restore Uploads: Copy/rsync backed-up uploads/data dir to the appropriate location.
3. Restore Config: Apply configuration envs/secrets; redeploy containers/pods.
4. Validate: Start platform, check flows, logs, and run test executions.

---

Disaster Scenarios & Recovery

• Full Cloud Outage: Bring up platform in alternate region/cloud using backed-up DB, uploads, and secrets.
• Accidental Data Deletion: Recover from backup taken hours ago; notify users of affected data.
• Admin/Config Loss: Use secured admin backup keys/configs to regain access.

---

Best Practices

• Rotate and test backups at least weekly. Log/test all restores.
• Encrypt backups in transit and at rest.
• Limit backup access to admins with audit logs.
• Maintain at least 2 geographically separate copies (for compliance, disaster recovery).
• Document and automate: Have a runbook with restore steps and contacts.

---

Reliable backup and tested restore is the foundation of business continuity—never deploy to production without validating your configuration!