Skip to content

Preserve and Release a Network-Lab Environment

  • Purpose: Turn compute release into an evidence-backed environment transition.
  • Owner: Platform operations
  • Impact: Stops the managed lab, writes recovery artifacts, and removes disposable resources only after verification passes.
  • Rollback strategy: If preservation fails, leave the environment retained, correct the failure, and rerun the export.

Pre-checks

Confirm the intended environment and inspect current state:

ENV=<environment>
REF=<blueprint-ref>
hyops show env --env "$ENV"
hyops blueprint preflight --env "$ENV" --ref "$REF"

For EVE-NG QEMU node-state capture, shut down stateful guests inside their operating systems and use the quiescence procedure before continuing.

Execute

Run protected release:

hyops blueprint destroy \
  --env "$ENV" \
  --ref "$REF" \
  --execute \
  --archive-before-destroy

Follow the exact confirmation shown by the CLI. Use --guest-quiesced for an EVE-NG operation only after the guest shutdown condition is true:

hyops blueprint destroy \
  --env "$ENV" \
  --ref "$REF" \
  --execute \
  --archive-before-destroy \
  --guest-quiesced

Verify

The command must report recovery verification before resource destruction. Confirm the final blueprint status is ok, preservation is verified, and the ongoing cost estimate is zero or reflects only intentionally retained reusable resources.

Records

Keep the blueprint destroy run record, recovery manifest, and any companion state records as one release decision. The record must identify both the verified recovery generation and the resulting infrastructure state.

Rollback

There is no destructive rollback after release. Reconstruct the environment through hyops blueprint deploy --restore-labs. If verification fails before release, the runtime retains the environment by design.

References