DEVOPSdigest

DevOps may not yet be perceived as a glamorous role (trust me, that time will come), but it performs a critical function in a cloud-based business: Keeping release cycles on track and revenue flowing.

DevOps must deal with zombies popping up on all sides, in the form of 500s, 404s, service timeouts, and fatal exceptions. Failing to attack these issues as soon as they crop up can generate rot throughout your business because customers have very little tolerance for a bad experience online. I’m a big believer in DevOps as the new face of customer experience.

Statistics tell us that no amount of testing will uncover every zombie lurking in your systems, especially the low-probability ones. People try to do things that even the best use cases don’t anticipate, or best QA teams every test for.

Therefore, operating at scale also means finding and fixing problems on the fly. At the core it’s simple: Humans write code, humans make mistakes, applications inherit them, and scale exposes them.

Luckily, you have a powerful zombie-killing weapon at your disposal: log management. By proactively monitoring for errors and performance degradation and allowing a log management solution to notify you through alerts, you’ll never have to sleep in fear again.

So as you load up your bag with treats on Halloween, be sure to load up on your #DevOpsZombieKillers before the next APP-pocalypse strikes. Here’s how:

■ Aggregate all of your log data with a convenient, cloud-based log management solution.

■ Set up dashboards and alerts to monitor your logs for errors and performance problems, so that you can detect the zombie incursion within minutes.

■ Have processes in place to troubleshoot operational issues and bugs in your code.

Take control of your fate: With these steps in place, you’ll rest easier. Without them, your site and apps might soon rest in peace.

Karen Sowa is a Marketing Specialist at Loggly.