Can a release be dropped soonish? Desire to use this with RKE 1.19.6...

[jchandler-parsons]

I'm working on an air-gapped system running RKE 1.19.6. We have a need to create, status and delete deployments at runtime from a Java application. This package seems to be the general answer to our needs, but the latest release (v2.8.2.1.1.1) uses deprecated APIs (deprecated as of K8s 1.16.x if I'm not mistaken). This prevents our application from successfully installing Tiller and providing any of the functionality we are looking for. There are a number of commits merged to the master branch that look like they should solve our Tiller install issue at least, and there is a PR for a Helm 2.16.2 equivalence that is over two years old. If I could convince our security folks to let me take the source code into our environment I would, but to follow their guidelines I have to have an official release from the package author(s).

From other issues/discussions, it appears that there is no desire to allow external help to aid the project/package development effort, but as far as I can tell, this is the only Java Helm binding utility out there, and it hasn't released in four years. It is open source, and I could always make a fork, but I'd rather let the original creator have a flourishing project if at all possible.

[ljnelson]

I'm sorry, I cannot promise anything. This was a hobby project that took off. Do note that Helm in the meantime has moved to an entirely different architecture in version 3 and Tiller is no longer required (which is a good thing). I will see if in my spare time I can release something but there are no guarantees.

[jchandler-parsons]

Are there known issues in the commits that have been merged to master that make them a bad candidate for release? The latest in master at least seems to have what is needed to get past the API deprecation issue I've specifically run into (TillerInstaller).

I understand that you have this project as a hobby, and it may have spiraled past what you believe you have the ability to support. The good news is that the project appears to fit a valid need (solves a problem rather than being a solution in search of a problem), and several people have made offers to help. That is one of the finer points/most exciting aspects of open source development. You don't have to do it alone. I have not looked at any of the forked projects, but the fact that you have so many people forking your project is another good indicator that it could be something that thrives, and if you don't have time to nurture it and own it, there are many people willing to help and let you take whatever role or commitment level you feel comfortable with.