Disaster-recovery runbook
Quick start

OneView documentation

Disaster-recovery runbook

Restore a coordinated client recovery set safely.

Recovery ownership

The client owns database backups, protected installation backups, replacement infrastructure, DNS, certificates, and the recovery decision. Assign a database restore owner, host restore owner, network owner, OneView validator, and incident decision-maker before recovery begins.

Use a compatible database and protected installation backup as one coordinated recovery set. Do not mix recovery points merely to make the application start.

Stabilize

  1. Isolate the failed or untrusted host and stop integration traffic.
  2. Preserve evidence according to client incident policy.
  3. Prevent the same installation from running on another host.
  4. Select a known compatible recovery point and record its OneView version.
  5. Confirm the required database, installation, certificate, and secret-recovery owners are available.

If only generated settings or permissions appear damaged and the host remains trusted, begin with Installation diagnostics & repair.

Restore

  1. Prepare a supported host with the required architecture, time, storage, and network access.
  2. Keep the failed host isolated.
  3. Restore the complete protected installation backup using the client's approved secret-handling process.
  4. Restore the matching database recovery point and verify database availability and backup integrity.
  5. Use the approved OneView recovery/start procedure for the exact recovered release.
  6. Generate a sanitized installation report.
  7. Restore and validate the approved HTTPS address.

Do not run initial setup over the restored installation, substitute a different software version, fabricate a missing installation identity, or start the same restored installation on two hosts.

Validate before return to service

  • Confirm the expected version and healthy application status.
  • Verify database connectivity and the recovered data point.
  • Open the approved HTTPS URL and run network checks.
  • Sign in with a named administrator.
  • Review and revoke restored sessions when the backup age or incident scope makes them untrusted.
  • Confirm license state, roles, service accounts, branding, vendor connections, and activity history.
  • Rotate credentials affected by the incident.
  • Run approved synthetic verifications.
  • Reconcile requests accepted after the selected recovery point.

Escalation evidence

Provide only the sanitized report, expected version and architecture, recovery-point timestamps, safe health facts, failure stage, and the smallest client-reviewed redacted log excerpt.

Never send database exports, unrestricted logs, configuration contents, credentials, setup or activation values, certificate private material, vendor credentials, or KYC data.

Return-to-service checklist

  • Failed host isolated and duplicate operation prevented.
  • Compatible database and protected installation recovery points selected.
  • Database, host, network, validation, and incident owners assigned.
  • Exact recovered version and approved recovery procedure confirmed.
  • Coordinated recovery set restored.
  • Application, database, HTTPS, license, access, activity, and synthetic checks passed.
  • Restored sessions and affected credentials reviewed.
  • Data-loss or reconciliation window recorded.
  • Return to service jointly approved.