Our serverless superhero this week is Chief Tobenna Nwokike, head of technology at iRechargeNG. Tobenna is highly active in serverless tech Twitter, constantly sharing with others and helping out when people raise questions. It’s this kind of engagement that really makes the community feel special. He is also building a serverless community in Nigeria, and will launch the site serverless.ng late this year. Thank you for all your contributions, Tobenna!
Matt Martz walks us through creating a cross-account IAM authorized API GW with the CDK this week. This step-by-step guide details everything you need to know to get this pattern working with the CDK. I can tell you from experience how difficult this is to figure out, so seeing this guide is a pleasant surprise!
Part of the reliability pillar in the serverless Well Architected framework from AWS is building your applications for retry. This is easier said than done. This week, I (Allen Helton) wrote about three levels of abstraction for retrying failures in your serverless apps. From code-based to generic across your AWS account, you have options with varying pros and cons.
Do you make use of Step Functions in your apps? Have you ever wondered what happens to your running state machines when you update the definition mid-run? Pubudu Jayawardana has done the heavy lifting and gotten to the bottom of it. He talks about what happens when you update the state machine definition vs when you update Lambda functions contained in the definition. Very interesting read!
A question that comes up regularly in development is “how do I schedule a job to run?” Adele Gauvrit tells us about a serverless solution to just-in-time scheduling she wrote. She has come up with a way to schedule events with a priority queue if necessary using the CDK. Clever design to a common problem!
When you practice CI/CD, you must be able to hide features in progress from production until they are ready. Feature flags are a great way to do this. Ran Isenberg gives us the details on not only how to use feature flags, but how to manage them like a boss. He talks about the lifecycle feature flags need to follow and how they should not be part of your codebase forever.
In case you missed it, DataDog recently released the state of serverless report. It was an interesting read, but blurs the lines a bit on what serverless actually means. Jason Wadsworth talks about his take on serverless and why the results of the state of serverless report should be taken with a grain of salt. Jason has the right idea here, we might talk more about this in the future.
Next week is the first day of summer. What better way to kick off the summer than to get involved in a serverless meetup group. Get to know some like-minded people in your area and learn a lesson or two from them. They are a great way to expand your skills and your social network.