Validating Backup Integrity For Recovery

Validating Backup Integrity for Recovery

When to Use

Use this skill when:

• Verifying backup integrity before relying on backups for ransomware recovery
• Building automated backup validation pipelines that run after each backup job
• Auditing backup infrastructure to confirm recoverability for compliance (SOC 2, ISO 27001, NIST CSF RC.RP-03)
• Detecting silent data corruption (bit rot) in backup storage before a disaster occurs
• Validating that immutable or air-gapped backups have not been tampered with

Do not use for initial backup configuration or scheduling. This skill focuses on post-backup validation.

Common Misconfigurations & Verification

• Archive integrity ≠ data integrity: a gzip -t pass or a valid restic check only proves the container is well-formed — the files inside can still be corrupted or pre-encrypted. Always hash individual files (SHA-256) against a baseline manifest, and run restic check --read-data / borg check --verify-data so the actual data blocks are read, not just the index.
• The most-missed failure is trusting a backup that predates detection but not infection: if the ransomware dwell time exceeds the backup age, your "clean" restore point already contains encrypted files or the implant. Scan restored data for ransomware extensions, ransom notes, and high file entropy (>7.9/8.0) before certifying it recoverable.
• Broken incremental chains fail silently: a single corrupted incremental invalidates every restore point after it, yet the backup job still reports success. Verify the full chain restores end-to-end, not just the latest snapshot.
• Weak/odd hashing and "immutable" that isn't: MD5 is broken — use SHA-256/SHA-3. And confirm immutability is real (S3 Object Lock, Azure immutable blob) by attempting a delete/overwrite and confirming it's denied, plus verifying air-gapped copies haven't been silently re-mounted to the network.
• Concrete verification it worked: restore to an isolated environment, then diff sorted baseline vs restored manifests (zero deltas), reconcile file counts and total size, run DB consistency checks (pg_restore --list, row counts), and confirm validation runs on a schedule with alerting on failure — a silently-logged failure is the same as no validation. Confirm the 3-2-1 layout actually holds (3 copies, 2 media, 1 offsite).

Prerequisites

• Access to backup storage (local, NAS, S3, Azure Blob, GCS)
• Python 3.9+ with hashlib (standard library)
• Backup manifests or baseline hash files for comparison
• Isolated restore environment for restore testing
• Backup tool CLI access (restic, borgbackup, rclone, or vendor-specific)

Workflow

Step 1: Generate Baseline Hash Manifest

Create a cryptographic fingerprint of every file at backup time:

# Generate SHA-256 manifest for a directory
find /data/production -type f -exec sha256sum {} \; > /manifests/prod_baseline_$(date +%Y%m%d).sha256

# Verify manifest format
head -5 /manifests/prod_baseline_20260319.sha256
# e3b0c44298fc1c149afbf4c8996fb924...  /data/production/config.yaml
# a7ffc6f8bf1ed76651c14756a061d662...  /data/production/database.sql

Step 2: Verify Backup Archive Integrity

Check that the backup archive itself is not corrupted:

# Restic: verify backup repository integrity
restic -r s3:s3.amazonaws.com/backup-bucket check --read-data

# Borg: verify backup archive
borg check --verify-data /backup/repo::archive-2026-03-19

# Tar with gzip: verify archive integrity
gzip -t backup_20260319.tar.gz && echo "Archive OK" || echo "Archive CORRUPTED"

# AWS S3: verify object checksums
aws s3api head-object --bucket backup-bucket --key daily/2026-03-19.tar.gz \
  --checksum-mode ENABLED

Step 3: Perform Restore Test to Isolated Environment

# Restore to isolated test directory
restic -r s3:s3.amazonaws.com/backup-bucket restore latest --target /restore-test/

# Generate hash manifest of restored data
find /restore-test -type f -exec sha256sum {} \; > /manifests/restored_$(date +%Y%m%d).sha256

# Compare baseline and restored manifests
diff <(sort /manifests/prod_baseline_20260319.sha256) \
     <(sort /manifests/restored_20260319.sha256)

Step 4: Validate Data Completeness

# Count files in original vs restored
echo "Original: $(find /data/production -type f | wc -l) files"
echo "Restored: $(find /restore-test -type f | wc -l) files"

# Check total size
echo "Original: $(du -sh /data/production | cut -f1)"
echo "Restored: $(du -sh /restore-test | cut -f1)"

# Database consistency check after restore
pg_restore --list backup.dump | wc -l  # Count objects in dump
psql -c "SELECT schemaname, tablename FROM pg_tables WHERE schemaname='public';" restored_db

Step 5: Detect Ransomware Artifacts in Backups

Before trusting a backup for recovery, scan for ransomware indicators:

# Check for common ransomware file extensions
find /restore-test -type f \( \
  -name "*.encrypted" -o -name "*.locked" -o -name "*.crypt" \
  -o -name "*.ransom" -o -name "*.pay" -o -name "*.wncry" \
  -o -name "*.cerber" -o -name "*.locky" -o -name "*.zepto" \
\) -print

# Check for ransom notes
find /restore-test -type f \( \
  -name "README_TO_DECRYPT*" -o -name "HOW_TO_RECOVER*" \
  -o -name "DECRYPT_INSTRUCTIONS*" -o -name "HELP_DECRYPT*" \
\) -print

# Check file entropy (high entropy = possible encryption)
# Files with entropy > 7.9 out of 8.0 are likely encrypted
python agent.py --entropy-scan /restore-test

Step 6: Automate and Schedule Validation

# cron-based validation schedule
# Run nightly after backup window
0 4 * * * /opt/backup-validator/agent.py --validate-latest --notify-on-failure
# Weekly full restore test
0 6 * * 0 /opt/backup-validator/agent.py --full-restore-test --config /etc/backup-validator/config.json

Key Concepts

Tools & Systems

Common Pitfalls

• Never testing restores: The most common failure mode. Backups that are never restored are untested assumptions.
• Checking only archive integrity, not data integrity: A valid tar.gz can contain corrupted file contents. Always hash individual files.
• Trusting last backup without scanning for ransomware: Backups may contain encrypted files if the infection predates the backup.
• Ignoring incremental chain integrity: A single corrupted incremental backup can break the entire restore chain.
• No alerting on validation failures: Backup validation must be monitored with alerts, not just logged silently.
• Using MD5 for integrity: MD5 is cryptographically broken. Use SHA-256 or SHA-3 for integrity verification.

References

• NIST SP 800-184: Guide for Cybersecurity Event Recovery
• NIST CSF 2.0 RC.RP-03: Backup Integrity Verification
• CIS Controls v8: Control 11 - Data Recovery
• CISA Ransomware Guide: https://www.cisa.gov/stopransomware