menu
save_alt Scarica

Disaster Recovery Plan

Recovery procedures for service outages and data loss

< Back

Overview

This plan outlines the recovery procedures for our services in the event of downtime, data loss, or platform outages.

1. Services & Infrastructure

1.1 Application

  • Platform: Heroku
  • Stack: Ruby on Rails

1.2 Database

  • Service: Heroku Postgres
  • Backups: Automated daily backups via Heroku

1.3 Website

  • Platform: Netlify
  • Type: Static site (e.g. marketing/docs)

2. Backup & Restore

2.1 Heroku Postgres

  • Backups: Daily via Heroku PGBackups + offsite backups to AWS S3
  • Retention: Based on Heroku plan tier; S3 backups per configured retention
  • Monitoring: Verify backup status monthly

Restore Procedure:


heroku pg:backups:restore '<backup_url>' DATABASE_URL --app <app-name>

2.2 App Code (Heroku)

  • Source of Truth: GitHub repository
  • Recovery: Redeploy latest stable version via Git

2.3 Website (Netlify)

  • Source of Truth: GitHub repository
  • Recovery: Redeploy via Netlify UI, CLI, or push to main branch

3. Failure Scenarios & Responses

3.1 Heroku App Down

  • Check: https://status.heroku.com
  • Inspect logs: heroku logs --tail
  • Redeploy if needed
  • If caused by a prolonged Heroku outage, communicate status to users and wait for platform recovery (Heroku manages all infrastructure)

3.2 Database Failure / Data Loss

  • Restore most recent daily backup
  • Validate on staging if possible
  • Communicate with users if downtime is required

3.3 Netlify Website Down

  • Check: https://www.netlifystatus.com
  • Re-deploy site from GitHub
  • If Netlify is completely down, wait for platform recovery or consider temporary alternative static hosting

4. Communication Plan

Internal

  • Notify team via Slack or email
  • Use incident channel if needed

External

  • Use status page, Twitter, or email list

Incident Template:

We’re currently experiencing downtime due to [brief reason]. The team is actively working on a resolution and will share updates shortly. Thank you for your patience.

5. Regular Checks

  • Check heroku pg:backups monthly
  • Test restoring backup to staging monthly
  • Review deploy logs and alerts regularly

6. Optional Improvements

  • Setup uptime monitoring (Pingdom, UptimeRobot)
  • Add fallback static hosting (e.g. alternative static host)