docker-lvmpy VS docker-py

Nodes in the SKALE Network are not required to have a specific amount of disk storage. There is a minimum viable quantity required, but not a set amount for a node. Variance in storage can create instability during some network functions. A collective decision was made in the community to fix the problem with software not hardware. A node could have X amount of storage, but is required to have Y amount of storage. Any excess in X over Y is now not utilized. This creates a uniform storage capacity amongst all nodes (assuming minimums are met) while not requiring the validator community to change any hardware specs. All while increasing stability and performance of the network and SKALE Chains.... ie this was a big Win!! You can see the code implemented here: https://github.com/skalenetwork/docker-lvmpy

In this specific example, I can't imagine how all of this exposed plumbing and error handling adds much value for the person reading or writing the code. In fairness, however, the Go code could easily be moved into function that does all of the dirty work and presents a minimal interface...which is what the Python example is doing:

https://github.com/docker/docker-py/blob/923e067dddc3d4b86e4...

Their Dockerfile seems quite straightforward. You could build the container yourself if you'd like.

Maybe try something like Docker SDK for Python (Probably other languages have SDK’s as well)

So, it turns out that docker-py does not (and possibly, will not) support Buildkit. Issue #2230.