Downtime is a reality for every business. The real question isn’t if your database will face an issue, but when. And when it happens, you have two choices: you can either panic while you figure out what’s connected and what the impact is, or you can rely on a well-prepared, step-by-step plan. 

So, invest time upfront in a solid business continuity plan. It’s what separates a minor disruption from a major business crisis, because it helps you minimise impact and get back online as quickly as possible. In this guide, we’ll walk you through the fundamentals of building a plan that actually works, based on our years of real-world experience with databases going down. 

What are you really preparing for? 

When people think about database risks, cyberattacks tends to be the first thing that pops into their minds. No big surprises there, since stories about dangerous viruses and hackers tend to generate great headlines. 

In reality, the threats are often much closer to home and far less dramatic. For example, human error is everywhere, but it tends to not be included in business continuity or disaster recovery plans. Think about what happens if a developer runs a faulty script on production or a system administrator pulls the wrong storage. 

Besides the human element, there are also external factors that you can’t control. A fire, a flood, or a power outage in your data centre can bring everything to a halt if you don’t have a recovery plan. It’s always best to plan for the worst. 

Ask the right questions first 

Of course, a good business continuity plan isn’t based on what feels right. You’ll have to answer some tough business questions to find your weak spots. Before you design any solution, you need to agree on three fundamental metrics: 

1. Recovery Time Objective (RTO)
   How long can this database realistically be offline before the business starts to seriously suffer? Be honest with yourself. Is it five minutes or five hours?
2. Recovery Point Objective (RPO)
   How much data can you afford to lose? This answer dictates how often you need backups, because losing the last hour of transactions is very different from losing the last 24 hours.
3. Uptime
   What is your total acceptable downtime? 99.9% uptime sounds great, but remember that it still allows for almost 9 hours of downtime per year. Make sure that’s acceptable to your business. 

These metrics will help to define your technical architecture, from standby nodes to a full disaster recovery site, but higher demands mean higher costs. The goal is to find the right balance for your business, not to chase an ideal that you don’t actually need. 

The golden rule 

One quick thing we want to point out before we move on: if you’ve never tested your business continuity plan, you don’t have one. You have a document. And it might work, but you can’t be sure.  

Regular, real-world testing is the only way to know for sure that your procedures hold up when the pressure is on. That’s why we like to say: an untested plan is a useless plan. 

Where to start: Two first steps 

If you want to get started with business continuity, the good news is that you don’t need to build a full-scale disaster recovery site tomorrow. Instead, you can start improving your resilience right now with a couple of low-effort but high-impact actions. 

First, get a regular health check. It’s not just important for people to stay on top of their health, databases need the occasional check as well! Having an expert review your infrastructure, performance and security every now and then will help spot problems long before they have a chance to become a crisis. 

Another great first step to take is to set up a standby node. Having a second database on standby and ready to take over can drastically cut your recovery time. This used to be an expensive, enterprise-level feature, but cost-effective tools like DBVisit bring this capability to standard editions. 

How we can help 

As experts in database recovery, we’ll gladly help you to build and maintain a business continuity plan. We’ll protect your business in three steps: 

1. First, we’ll sit down with you to review your existing plan or help you build one from scratch. We make sure it matches your real business needs, not just a generic template.
2. When an incident happens, our experts are on call with 24/7 support to help you execute your recovery and get your systems back online, fast.
3. And of course, the best way to handle a disaster is to prevent it. That’s why we’ll help you integrate monitoring tools to spot potential issues early and stop downtime before it starts.
    

Are you ready to stop reacting to downtime and start preparing for it? Get in touch with us to get expert advice on business continuity! 

P.S. If you’re using SQL Server, one great way of avoiding downtime is to use Distributed Availability Groups during migrations. Check out our full blog on DAGs for more info!