Data backup · Pinellas County

Disaster readiness for Pinellas sites.

Protect your site during storm season with planned backups, tested restores, and local escalation paths.

Backup strategy.

  • Daily full backups + hourly diffs for databases during business hours.
  • Retention tuned to your risk profile (30–90 days).
  • Offsite copies plus local oneColo storage for fast restores.
  • Encrypted at rest and in transit.
Recovery plan
  • Staging environment for validation before going live.
  • Documented RPO/RTO targets and runbooks.
  • Escalation to on-site oneColo support when needed.
  • Quarterly restore drills to verify backups.
Plan my backups

RPO and RTO that match business reality.

Define recovery objectives by system, not guesswork, so you know how much data and downtime you can actually absorb.

  • Recovery Point Objective (RPO): how much data loss is tolerable—e.g., 15 minutes for orders, 1 hour for content.
  • Recovery Time Objective (RTO): how fast systems must come back—e.g., 30 minutes for checkout, 4 hours for marketing pages.
  • Prioritized tiers: critical (checkout/auth), important (forms/CRM), and standard (blog/media), each with clear RPO/RTO.
  • Runbook alignment: who restores, in what order, with which validation steps, and when to fail over.
RPO / RTO Clear targets
  • Document per-system RPO/RTO with owners and escalation paths.
  • Test restores against those targets; adjust cadence if you miss them.
  • Keep staging in sync to validate before production cutovers.
  • Monitor backup health and alert on failures before an incident.
Set my RPO/RTO targets

Vibe check: bandwidth and seed planning.

Backups and restores are only as good as your uplink. Don’t wait to find out during an incident.

  • Uplink speeds: test real upload throughput; hotel or WFH Wi-Fi won’t cut it for large restores.
  • Initial seed backups: ship first copies over LAN or direct-connect before relying on offsite sync.
  • Restore windows: estimate time-to-restore over your uplink for databases, media, and full site images.
  • Warning signals: backup jobs overrunning their window, failed checksums, or nightly uploads that never finish.
Bandwidth reality Warning signs
  • If uplink is under 20–30 Mbps, plan seeded backups and staged restores.
  • Schedule backup windows outside peak hours; monitor duration creep.
  • Test restores quarterly over the same network you’ll use in an incident.
  • Document alternative paths (on-site handoff, drive swap, or temporary colo access) if uplink is saturated.
Check my backup readiness