The Future of Terraform and the Terraform Associate 003 Exam
Lots of news around Terraform lately. This could be THE time to get involved and up your Terraform game. Or, it might be a total waste of time.
I've been spending a lot of time with Terraform over the last couple years. I've become generally familiar with it, but the truth is that I've mainly been spending that time modifying code (infrastructure) that was already built and deployed. I'm adding new IAM policies here, a couple new metric there, a few variables — but, with job scope shifts on the horizon, I've reached a point that made me want to have a solid base of knowledge from which to work. Notably, I've been using a proprietary internal provider that just doesn't translate to real world AWS or GCP (etc etc) experience.
I've talked about this before — all the time I spend googling answers (or searching slack and checking confluence) definitely get's the job done, but I lose out on all the benefits that come from really knowing something. And sometimes, you're in an interview and they ask you a question ... and realize that you can't answer because you never truly worked it out in your own head. And boy do they know as you stumble over your words trying not to sound like an idiot.
Anyway, it's a weird time to invest in Terraform. Hashicorp has moved Terraform from an open-source license to a more restricted BSL license. This won't affect my job (for now) since we're only using a custom provider, but this could be a reason to shift away from Terraform over the next months/years. Once I had heard the news, I immediately started looking for Terraform alternatives... almost instinctually. But, the landscape has changed so much since the end of the pandemic. This might be the time for Hashicorp to do this. It is, after all, entirely reasonable for a company to seek profit. I just hope they're ready for the competition.
The Terraform Associate 003 exam
I have a weird relationship with these certifications. I really don't expect them to change my life or open brand new doors that lead to giant stacks of cash. But, I can't deny that they're a great way build a thorough knowledge base.
Sure, they only ask you 60 questions and you have to wade through a hundred pages in a library of information to secure a certified terraform engineer “badge”. But, the benefit is actually hidden right there in all that information that you wade through. You might not need to know it all, but it's not going to hurt you.
The good news about the exam is that it focuses primarily on practical knowledge of the actual language, and I think you can pass just knowing the ins and out of HCL.
Terraform is all about modularization and reusing code, both in the real world and in the exam. If you're using it for your own projects you might not have invested time in writing modules, but I can tell you that they're essential as your infrastructure expands. Make sure you know how to take advantage of them. Write a couple and use a couple yourself.
Use a few different cloud providers and their terraform providers. AWS is the defacto place to start, but try OCI if you want a challenge.
Get to know the Terraform CLI. Not much to it, but use it to generate a few plans and an apply or two. Helps to know about validate and fmt as well.
Finally, it's only $70 so maybe you can convince your boss to reimburse it if you pass.
The bad news is that Hashicorp uses it somewhat as a marketing platform to push their own tools like Terraform Cloud and Sentinel. Don't get me wrong, they're neat tools that I wouldn't have known about had I not required myself to pass this test, but knowing about them is less likely to help you in the real world.
See something wrong? Email [email protected].