Disaster Recovery Planning for Servers: RTO, RPO and a Step-by-Step Plan

Disaster Recovery Servers

Disaster recovery (DR) planning is how you make sure mission-critical servers and data come back quickly after a failure, theft, fire, ransomware attack or extended power outage. A good plan is not a thick document nobody reads. It is a clear set of targets, systems and steps that you have actually tested.

Quick answer: a disaster recovery plan sets a recovery time objective (RTO) and recovery point objective (RPO) for each critical system, uses backups and replication that meet those targets, keeps at least one copy of data off site, documents the recovery steps, and is tested regularly. Start with your most critical systems and build out from there.

The key terms

TermMeaningExample
RTO (recovery time objective)How long a system can be downAccounting system back within 4 hours
RPO (recovery point objective)How much recent data you can afford to loseNo more than 1 hour of transactions
BackupA copy of data you can restoreNightly backup to a backup server
ReplicationContinuously copying systems or data to another locationVMs replicated to a standby host
FailoverSwitching to a standby systemStarting replicas at a second site
Virtualization in Disaster Recovery

Step 1: Identify critical systems

List every system, then rank them by business impact. Typical tier 1 systems are finance and ERP, email, customer-facing applications and file shares that staff need daily. Set an RTO and RPO for each tier.

TierTypical RTOTypical RPOApproach
Tier 1 (critical)Minutes to hoursMinutesReplication to a standby server or second site
Tier 2 (important)Hours to a dayHoursFrequent backups and a ready restore target
Tier 3 (non-critical)DaysA dayNightly backups
Good to know: these tiers are a starting framework. The right targets come from asking what an hour or a day of downtime would actually cost your business.

Step 2: Choose your recovery methods

  • Backups following the 3-2-1 rule: three copies, on two types of storage, with one off site.
  • Immutable or offline backups so ransomware cannot encrypt or delete them.
  • VM replication to a standby host for systems with short RTOs. See setting up virtualisation.
  • A recovery site: a second office, a data centre or cloud recovery service.
  • Power protection with UPS and generator or inverter capacity for load shedding.
Pro tip: refurbished servers are a cost-effective choice for standby hosts and backup targets, because they spend most of their life idle. Read how refurbished servers fit into disaster recovery plans.

Step 3: Document the plan

  1. Contact list: who declares a disaster, who does what, and supplier contacts.
  2. System inventory with RTO, RPO and recovery method for each.
  3. Step-by-step recovery procedures, written so someone else can follow them.
  4. Locations of backups, passwords and licence keys, stored securely off site.
  5. Communication plan for staff, customers and, where personal information is affected, the Information Regulator.
Watch out: keep a copy of the recovery plan, passwords and licence keys outside the systems you are trying to recover. A plan stored only on the server that failed is no help.

Step 4: Test and maintain

Test at least once or twice a year: restore complete servers, not just files, and time how long it takes against your RTO. Update the plan after every test and every major change. Pair this with high availability design to reduce how often you need it.

Frequently asked questions

What is a disaster recovery plan?

A documented, tested set of targets, systems and procedures for restoring critical IT services after a major failure or incident.

What is the difference between RTO and RPO?

RTO is how quickly a system must be restored. RPO is how much recent data you can afford to lose.

How often should disaster recovery be tested?

At least once or twice a year, and after significant changes to your systems.

What is the 3-2-1 backup rule?

Keep three copies of your data, on two different types of storage, with one copy off site.

Building a standby or backup server? Browse our storage servers or talk to our team.

Server Site