Operator Guide
Deploy a recoverable MP-OPT service
Follow this sequence from top to bottom. The first four stages establish a safe standalone server. High availability is an optional extension, not a substitute for recovery.
Supported operator path
A recoverable first server is the common foundation. High availability is added only after that baseline works.
Step 1
Plan
Choose standalone now, with optional HA later
Step 2
Deploy
Install and bootstrap the first VPS
Step 3
Recover
Create, export, and verify an encrypted snapshot
Step 4
Commission
Prove desktop publish and web access
Step 5
Add HA
Optional second VPS, witness, routing, and tests
Step 6
Operate
Monitor, update, practise, and recover
Operator safety rule
Do not admit important data until an external recovery snapshot has been exported and deep-verified. Do not enable automatic failover until the HA commissioning sequence passes in both directions.
Step-by-step outline
1. Choose and prepare
Done when: You know the deployment model, hostname, provider responsibilities, and where recovery material will live.
Choose a deployment2. Establish the first server
Done when: One VPS serves HTTPS, the root passkey works, and the guarded management console reports healthy.
Install and bootstrap3. Prove recovery
Done when: An encrypted, deep-verified recovery snapshot and its separately protected private identity exist away from the VPS.
Protect recovery material4. Commission standalone operation
Done when: Desktop publishing, passkey access, web viewing, snapshot export, and restore prerequisites are verified.
Run standalone verification5. Optionally add HA
Done when: A second VPS can safely take ownership, and real failover has passed in both directions.
Understand and configure HA6. Operate and recover
Done when: Routine checks, upgrades, failover, and incident decisions follow documented guarded workflows.
Open day-to-day operations
Three independent resilience layers
Availability, recent redundancy, and disaster recovery solve different problems and must not be treated as interchangeable.
1. Live service
The holder serves requests. In HA, either VPS may hold the writer lease.
2. Peer recovery point
A recent accepted complete bundle supports failover after one VPS is lost.
3. External recovery copy
An independently encrypted and verified snapshot protects against both VPSs, corruption, and operator mistakes.