As time marches on, so does technology. Either you adopt to the changes as they approach, or wake up to A Bad Day at work. That's what happened to one of my customers way back in 2013 when they found they could not connect RDP to their fleet of >1000 Windows instances because they were not updating the images, even though the breaking change was communicated for 3 years prior. Restarting the instances with a user data Powershell script to update the instances fixed the issue, but it took a few days of development and testing before services were back online.
Don't have a bad day and take a look at your stack for possible deprecation!
Here are three examples that you may want to pay attention to if you work on AWS:
The Amazon EC2 Instance Metadata Service will be switching to IMDSv2 for new instances this summer. You can read more about that in this blog post. While you can still turn on v1, there are good security reasons to migrate. Read the docs on how to transition from v1 to v2 today!
The AWS Go SDK v1 will enter maintenance mode and be unsupported effective June 31, 2024. Time to update all those Terraform scripts!
While Amazon Linux 2 End-Of-Life was extended to 2025, you should really start migrating to AL2023 as soon as possible. There are lot of fundamental differences that have a high chance of disrupting your applications if you do not test them. Read more about that the differences in the AL2 and AL2023 in the documentation and make plans to migrate within the next 6 months.
These are just a couple of the examples I had conversations about in the last few weeks, there are more. It's worth repeating:
Don't have a bad day and take a look at your stack for possible deprecation!
Top comments (0)