Okay, lets talk nerdy. One of most fundamental underpinnings of the way we think about doing infrastructure automation is the idea that it's best done in a declarative way. Having a declaration of intent, followed by an idempotent, convergent control loop.
1/ A world first reverse engineering analysis of AWS Session Tokens.
Prior to our research these tokens were a complete black box. Today, we are making it more of a glass box, by sharing code and tools to analyze and modify AWS Session Tokens.
https://t.co/a1d4iznkSs
PSA: if you're using CDK with esbuild to bundle your Node.js/TypeScript #lambda functions, pay attention to the last 0.22 esbuild update. The default behavior has changed and it will no longer bundle your dependencies.
If you want to restore the default, use the config below:
@badlogicgames@unsereOEBB Man kann Reservierungen bei der ÖBB leider nicht in der App stornieren.
Habe mindestens 2 Reservierungen von denen ich weiß das ich nicht da sein werde, aber keinen Weg gefunden das der ÖBB mitzuteilen
It's like how almost every list of company values contains the exact same bullshit about "Excellence, Hard Work, Integrity, Customers First, Teamwork, Passion." Gross.
But each company has its own unique character, culture, reputation, conventions. Vibes. That shit is real.
I'm not going to call anyone out, but I have been 😳 at the many people who replied with some variation of, "migrations are never worth it."
Kids.
If you aren't migrating, you are dying.
If it hurts, do it more.
Hey @purtelnoc your DNS resolvers seem to cache NS records for much longer than 2 days, I guess you want to check that, to avoid issues for your customers
For most of my career, I have been on call for the systems I built.
This experience forged my strong affinity for technologies that have less runtime complexity. That's why I'm so into serverless.
Assuming that complexity cannot be destroyed and can only be moved around. I much prefer technologies with less operational complexity (think patching AMIs, updating RDS versions, replacing failed nodes, etc.), even if it comes at the expense of higher author-time complexity.
That's why I prefer DynamoDB over RDS. Even though DynamoDB is rather restrictive and makes many things hard to implement.
That's why I prefer Lambda over running containers. Even though running applications in containers still provides a better developer experience.
My choices are not absolute. It always depends on the context.
And what I'd do myself is often different from what I'd advise my clients. Because they have different skill sets and must operate within some time and budget constraints.
But if it's a system that I need to maintain, then there must be a lot of upside to tip me over the fence.
Of course, your mileage might differ, but that's just me.
@LarsFronius In the depths of AWS docs, there is a description how to request a reverse DNS record. I used the same way to ask them to remove it, and it actually worked.
#AWS Going great again...
Want to delete a Lightsail static IP. Error Message: Can't delete. Contact Support. Trying to contact support: Technical Support only in paid plans.
So either I pay support, or I pay for the unused IP
"How can I put developers on call for their code, but prevent them from getting paged about other people's code or problems with shared resources?"
My sweet kittens, you cannot. Not without a separate database.
It constantly surprises me how widely misunderstood this is.
🚀 Ready for a deep dive into CDK with Andreas Sieferlinger? 🌐 Explore the tradeoffs in developing CDK constructs and libraries, and uncover the best options for your projects! 💡🛠️ Don't miss out on this invaluable insight. Let's level up our CDK game! 👨💻👩💻 #AWSCommunityDayDACH
🛠️ Dive deep into CDK development with @webratz! Compare tradeoffs, project setup, and custom checks for CDK constructs and libraries. Choose the best option for your projects! 💡 #AWSCommunityDay#AWSCommunityDayDACH Register: https://t.co/NB2mYsnZgP