init-snapshot VS aws-codebuild-docker-images

At CodeSandbox we use Firecracker to run our VMs (more info here: https://codesandbox.io/blog/how-we-clone-a-running-vm-in-2-s...).

To answer the questions:

> what version of the kernel do you use (the github page says 5.10 but isn't that quite old?)

Right, they have tested with 5.10, but it also works with higher kernel versions. Our host currently runs 5.19 and we're planning to upgrade to 6.1 soon. The guest runs 5.15.63, we use a config very similar to the recommended config by FC team (it's in the FC repo). It's important to mention that we had to disable async pagefaulting (a KVM feature) with more modern kernel versions, as VMs could get stuck waiting for an PF resolve.

> What do you use to build the 'micro' images

We created a CLI that creates a rootfs from a Docker image. It pulls the image, creates a container and then extracts the fs from it to an ext4 disk. For the init, we forked the open sourced init from the Fly team (https://github.com/superfly/init-snapshot) and changed/added some functionality.

> How do you keep timesync of you're not using a timesync daemon?

IIRC we expose the time as a PTP device (handled by kvm) and run phc2sys to sync the time in an interval. Firecracker has some documentation on this, where it recommends chrony. It can also be done with vsock, but it would be more manual.

> Handle kernel and app logs without adding an log daemon, and same through vsocks, etc?

The init forwards stdout/stderr of the command it runs to its own stdout, which Firecracker then logs out by itself. A supervisor reads these and writes the logs to files.

Unless they’ve changed things, there is no containerization within the VM a la kata. They run their own custom init inside the VM and use it to start the entry point. https://github.com/superfly/init-snapshot is the source.

Jerome wrote our init in Rust, and, after being cajoled by Josh Triplett, [we released the code (https://github.com/superfly/init-snapshot), which you can go read.

> Fly.io transforms container images into fleets of micro-VMs running around the world on our hardware.

Oh boy!

> None of us have ever worked for Google, let alone as SREs. So we’re going out on a limb

Oh.... boy.

> We spent some time scaling it with Thanos, and Thanos was a lot, as far as ops hassle goes.

You know, they have these companies now, that will collect your metrics for you, so that you don't have to deal with ops hassle.

Holy shit. I see they even wrote their own init... in Rust. Yes, the thing that is normally a shell script, is now a compiled program in a new language, that mostly just runs mkdir(), mount() and ethtool(). (https://github.com/superfly/init-snapshot/blob/public/src/bi...)

Based on public repository for Codebuild Image, the image base will be the Ubuntu standard 7.0.

This is basically what CodeBuild does.

The default Docker containers that CodeBuild uses (you can create your own) and the shell script it uses to parse the yaml configuration file (mostly a list of shell scripts) are all open source and the entire process can be run locally.

https://github.com/aws/aws-codebuild-docker-images

https://docs.aws.amazon.com/codebuild/latest/userguide/use-c...

Disclaimer: I work for AWS. But nowhere near the team that developed Firecracker

Disclaimer: I work for AWS in Professional Services. All opinions are my own.

The beauty about CodeBuild is that there is no “lock-in”. All it is fundamentally is a Linux or Windows Docker container with popular language runtimes and a shell script that processes a yaml file or you can supply your own Docker container.

You just put a bunch of bash commands or PowerShell commands in the yaml file and it runs anything.

The Docker container and the shell scripts are all open source and you can quite easily run them locally.

I could see outside of AWS keeping your Docker containers for your specific build environments in a local repository and doing all of your builds inside them using Jenkins.

https://github.com/aws/aws-codebuild-docker-images

https://docs.aws.amazon.com/codebuild/latest/userguide/use-c...

For a “batteries included” approach though, I really like Azure DevOps Pipelines.

I’ve even done a couple of integrations between Azure DevOps and AWS when we had clients that are Microsoft shops.

https://aws.amazon.com/vsts/

For AWS, if you use CodeCommit (AWS git service), all access is via IAM and granular permissions. If you integrate with Azure DevOps, the AWS credentials do have to be stored in a separate MS hosted credential storage.

CodeBuild also supports at least Github natively.

I’m not shilling for AWS. I have an MS development background (.Net) and only have “DevOps” experience using AWS and Microsoft tooling.

Docker Images provided by the CodeBuild team should be updated regularly and should support all "modern" toolkits. The open source project has some activity, but an issue for supporting newer Android versions is now open for some time...

The runtimes available and exposed by CodePipeline support Android runtime 29 – and the Docker images are provisioned using Java 8. Unfortunately, as of July 2021, the Android gradle tools (used by Flutter) require Java 11. I have created an issue in the corresponding Github (see here) but needed to find a workaround to move on – I think I’ve found one, but I hope that anyone reading this might have a better way or idea?

$ git clone https://github.com/aws/aws-codebuild-docker-images.git $ cd aws-codebuild-docker-images/al2/aarch64/standard/2.0 $ docker build -t codebuild/amazonlinux2-aarch64-standard:2.0 .

This also is the place where we specify this is an AArch64 build. The managed image indicates to use a standard image provided by AWS. The source of the Graviton2 image can be found on GitHub.

🏗 AWS CodeBuild Local Builds - Simulate a CodeBuild environment locally to quickly troubleshoot the commands and settings located in the BuildSpec file.