dinker VS nix2container

(Self plug) I had the same thoughts as the author, and made this: https://github.com/andrewbaxter/dinker . Like stated in the article, if you're doing rust or go all you want is to dump the binary in the image. There's no reason to do the build inside the docker vm in that case, and it's super fast, and only uses dumb filesystem access - no daemons like docker, weird wip container managers like buildah, etc.

Self hoisting here, I put this together to make it easier to generate single (extra) layer docker images without needing a docker agent, capabilities, chroot, etc: https://github.com/andrewbaxter/dinker

Caveat: it doesn't work on Fly.io. They seem to be having some issue with OCI manifests: https://github.com/containers/skopeo/issues/1881 . They're also having issues with new docker versions pushing from CI: https://community.fly.io/t/deploying-to-fly-via-github-actio... ... the timing of this post seems weird.

FWIW the article says

> create a Docker image, also known as an OCI image

I don't think this is quite right. From my investigation, Docker and OCI images are basically content addressed trees, starting with a root manifest that points to other files and their hashes (root -> images -> layers -> layer configs + files). The OCI manifests and configs are separate to Docker manifests and configs and basically Docker will support both side by side.

No discussion about Nix-built containers is complete with nix2container:

https://github.com/nlewo/nix2container

It is truly magical for handling large, multi-layered containers. Instead of building the container archives themselves and storing them in the nix store, it builds a JSON manifest that is consumed by a lightly patched version of skopeo that streams the layers directly to either your local container engine or the registry.

This means you never rebuild or reupload a container layer that is unchanged.

- Container images are built with https://github.com/nlewo/nix2container instead of Nix's built-in docker tools, because nix2container allows configuration of layers and thus makes it easier to share layers between images and improve layer caching.

As for integrating it into our CI pipeline, this is pretty much CI system agnostic as all you have to do is run `nix build ...` inside the pipeline, generating a `result` directory or file as an artifact that, in the case of nix2container, can be pushed to your container image registry.

dockerTools is the canonical way, but https://github.com/nlewo/nix2container seems to be gaining support as it's faster and doesn't use so much disk space while building.

> The key factor behind our decision was the realization that while Docker images are industry standard, moving around 100s of megabytes of images seems unnecessarily heavy-handed when we just need to synchronize a small change.

I think the culprit is more the GitHub Actions cache than Docker since it seems to be hard to get a clean cache management. I'm not sure about caching Docker image layers, but caching the Nix store with GitHub Actions is pretty complicated (not even sure it's possible): this means we have to download all required Nix store paths on each run, but i consider this is because of a GitHub Action cache limitation.

So, did you consider using another CI, which offers better caching mechanisms?

With a CI able to preserve the Nix store (Hydra[1] or Hercules[2] for instance), I think nix2container (author here) could also fit almost all of your requirements ("composability", reproducibility, isolation) and maybe provide better performances because it is able to split your application into several layers [2][3].

Note i'm pretty sure a lot of Docker CI also allows to efficiently build Docker images.

[1] https://hercules-ci.com/

[2] https://grahamc.com/blog/nix-and-layered-docker-images

[3] https://github.com/nlewo/nix2container/blob/85670cab354f7df6...

FYI, the nix2container [1] project (author here) aims to speedup the standard Nix container workflow (dockerTools.buildImage) by basically skipping the tarball step: it directly streams non already pushed layers.

[1] https://github.com/nlewo/nix2container

To get Rust incremental builds, did you consider using something such as crane https://github.com/ipetkov/crane ?

And regarding OCI images, i built nix2container (https://github.com/nlewo/nix2container) to speed up image build and push times.

For a more modern iteration of that idea, check out https://github.com/nlewo/nix2container

It has all the benefits of OP but produces the images much faster, and with less on-disk usage.

I've been using nix2container[1] for awhile now. It looks like this depends on Docker, have you guys considered removing that dependency? It shouldn't be necessary to create an OCI compliant image with Nix.

[1]: https://github.com/nlewo/nix2container

With https://github.com/nlewo/nix2container, I'm trying to make a more standalone tool. Basically, a Go binary takes a reference graph and produces a JSON file describing a container image. This JSON file is then ingested by a Skopeo fork (it adds a new `transport`) to produce images (to file, registries,...).

Currently, it supports the dockerTools layering algorithm and is designed to work with Guix [1] as well;)

[1] https://github.com/nlewo/nix2container/blob/065e5b108650ee4c...