We’ve blogged before about the importance of limiting the chances of vendors being single points of failure for your organization.  Avoiding such instances should be an important part of your uptime strategy.  Over 6,000 canceled flights by Delta Air Lines as a result of the ill-fated CrowdStrike software update is a good (bad?) reminder of the concept. 

Delta’s situation is that CrowdStrike’s faulty software update was able to bring down not just the same systems as many of their competitors, but Delta’s pilot and flight attendant scheduling as well.  While other airlines recovered, Delta struggled for days.  Even passengers with flights not cancelled but merely delayed for hours heard announcements over the airport PA system looking for a single available flight attendant so they could be on their way.  The entire episode was painful for passengers, their families and business colleagues, other crew members, and everyone in the travel ecosystem. 

One basic data center strategy for many organizations is to have a small disaster recovery presence in a colocation data center.  In addition to serving as failover, when architected correctly it can allow you to “turn back the clock” to before an outage happened, accelerating recovery while limiting the time period of data loss. 

We’re not suggesting this simple strategy would have worked for Delta in this instance, but it may be worth your consideration to enhance your uptime posture against both outage and ransomware. 

For more information on avoiding downtime, whether via hacker or vendor-fueled outage, read the Direct LTx whitepaper Maintaining High Availability:  Nine Critical Steps to Take for Disaster Recovery Success.   Click here to read the whitepaper or contact us at strategy@directLTX.com with questions or to arrange for a data center tour.