Skip to content

Disaster Recovery

Disaster recovery planning

How quickly you come back up after an outage — and how much data loss you can accept — is decided in advance. Scenarios built around RTO and RPO targets are rehearsed regularly.

Recovery time objective
RTO
Recovery point objective
RPO
Scenario verification
Rehearsal
Secondary copy location
Separate site

RTO and RPO: the plan starts with two questions

A disaster recovery plan starts not with a technical preference but with two business decisions. First, RTO: how long can you stay down when the system fails? Second, RPO: how far back in time can you accept losing data?

Investment made without settling these two numbers ends up either more expensive than necessary or insufficient. On a system backed up once a day the RPO is at worst 24 hours; an e-commerce operation that cannot tolerate that needs replication. Conversely, targeting a minutes-level RPO for an internal reporting server is wasted cost.

Once the targets are clear the scenario is designed: which systems come up in which order, how DNS and routing are switched, which source data is restored from, who is informed and when. Not having these steps written down is the single biggest source of lost time during a crisis.

The final and most frequently skipped part of the plan is rehearsal. A recovery plan that has never been exercised is a document assumed to work. We repeat the scenarios at regular intervals, compare measured durations against targets, and feed the results back into the plan.

  • RTO and RPO targets are documented
  • The recovery order of systems is defined in advance
  • A secondary copy is kept at a separate site
  • Rehearsal results are measured and fed back into the plan
  • Crisis communication and ownership are made explicit

Process

The planning process

  1. 01

    Criticality analysis

    Which systems hurt the business most when they stop? Systems are ranked by criticality.

  2. 02

    Target setting

    RTO and RPO values are agreed per critical system, weighing cost against risk.

  3. 03

    Scenario design

    Recovery steps, ordering, owners and the communication flow are documented.

  4. 04

    Rehearsal and revision

    The scenario is executed, durations are measured, and the plan is revised where it deviates from target.

What the plan must cover

  • The list of critical systems and their recovery order
  • An RTO and RPO target for each system
  • Which source data will be restored from
  • The steps for DNS, routing and access changes
  • Who owns which step
  • How customer and staff communication is handled
  • Failback steps for returning to normal
  • The rehearsal calendar and a record of results

FAQ

Frequently asked questions

I have backups — do I still need a DR plan?

Backup protects the data; disaster recovery plans how quickly the service returns. Even with backups, rebuilding the server, configuring it, restoring data and switching routing can take hours. If that duration is unacceptable, you need a plan.

How should I set RTO and RPO targets?

It is a commercial rather than a technical question: what does an hour of downtime cost you, and what does an hour of lost data cost? Once those two figures are clear, the sensible solution becomes obvious. We run the analysis with you.

Does a rehearsal affect production?

Rehearsals are run by restoring into a separate environment without touching production. The goal is to verify that the recovery steps work and the durations meet target, not to put production at risk.

Let's plan your infrastructure together

Tell us what you need and we prepare a configuration and pricing specific to you. We don't sell fixed packages; we build the deployment, licensing and backup plan with you.