agentleFS
Sign inStart free
mejbaurbaharmejbaurbahar/software-tester-skills/skills/disaster-recovery-backup-testing/SKILL.md

disaster-recovery-backup-testing

Skill1 starsChanged 55 days agoMITView on GitHub

Use when validating backup, restore, high availability and disaster recovery — RTO/RPO verification, restore drills, region failover, data corruption and ransomware recovery, runbooks, game days, and backup integrity checks for databases and storage.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    A backup you have never restored is a hope, not a backup: Test recovery against *stated objectives*.

    line 13

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Disaster Recovery & Backup Testing

A backup you have never restored is a hope, not a backup. Test recovery against stated objectives.

Define the targets first

  • RPO (max tolerable data loss, e.g. 5 min) → sets backup/replication frequency.
  • RTO (max downtime, e.g. 1 h) → sets automation and runbook needs.
  • MTD/priority tiers per service (Tier 0: auth/payments … Tier 3: internal tools).

Test ladder (do all, regularly)

LevelExerciseFrequency
1. Backup integrityJob success, size sanity, checksum, encryption, offsite/immutable copy existsdaily automated
2. Restore testRestore latest backup to isolated env; row counts/checksums; app boots and passes smokeweekly/monthly automated
3. Point-in-time recoveryRestore to T-15 min before a deliberate bad DELETE; verifyquarterly
4. Component failoverKill primary DB/cache/queue; replica promotes; app reconnects; measure RTO/RPOquarterly
5. Zone/region failoverTraffic shift, data replication lag, DNS TTL, secrets/KMS/IAM in DR region1–2× / yr
6. Game dayUnannounced-ish full scenario incl. people, comms, access, runbook1–2× / yr
7. Cyber recoveryRansomware: restore from immutable/air-gapped copies into clean roomyearly

Restore verification script (sketch)

pg_restore -d restore_test latest.dump
psql restore_test -Atc "select count(*) from orders" # compare with prod snapshot
psql restore_test -Atc "select max(updated_at) from orders"   # RPO evidence
./smoke.sh https://restore-test.internal                        # app-level check
# Point-in-time (Postgres WAL)
recovery_target_time = '2026-09-21 14:05:00+00'  ; recovery_target_action='promote'
aws rds restore-db-instance-to-point-in-time --source-db-instance-identifier prod --target-db-instance-identifier drill --restore-time 2026-09-21T14:05:00Z

What to verify beyond data

Secrets/KMS keys available in DR · IAM/roles · DNS/failover TTLs · TLS certs · third-party allowlists (IP changes) · config/flags · queues/streams offsets · search indexes & caches rebuilt · scheduled jobs not double-running · email/SMS callbacks · monitoring & alerting present in DR · on-call access without VPN dependency on the failed region · runbook is accurate, step-by-step, and owned.

Failure scenarios to script

Accidental delete/DROP · corrupt migration · bad deploy (rollback) · AZ/region outage · credential compromise · ransomware · vendor outage (auth, payments, DNS) · expired cert · disk full · clock skew.

Metrics & evidence

Measured RTO/RPO vs target, restore duration trend, backup success %, time to detect, time to decide, gaps found & closed; store evidence for audits (compliance-testing: SOC 2 A1.2, ISO 27001 A.8.13/A.5.30).

Related

chaos-resilience-testing, data-migration-testing, cloud-infrastructure-testing, observability-testing, release-readiness-testing

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free