Is elktail abandoned?

[dualbus]

The most recent commit (e350718) happened on 2017-11-07, almost a year ago. This project seems abandoned.

[getsomebread]

I love elktail and find it quite useful, I hope not!

[bergerx]

No comments for 4 months and i see no replies to any issues or PRs in last year. And PRs like #34 is sitting in a corner.

Maybe worth checking current forks if there is any other one currently maintaining elktail properly and migrate there.

@knes1 any comments? Would you consider moving the repo to a Github organization and let other volunteers to help maintaining it. I'm seeing some PRs coming already.

[knes1]

Hi everyone -

I developed elktail to solve an issue that I was facing (and still am) at the time - ability to quickly look at the logs stored in EL from command line. And elktail is fulfilling that for me - I use it on daily basis.

Now here's the issue with the PRs:

• I have very little time to dedicate to this project
• I've looked through the PRs and the forks and for many of them I ultimately decided against trying to merge them at this time.

Here's my perspective on this and some of the reasons:
a) They add a feature that I wouldn't know how to test in my environment, and therefore I don't know if/how it's really working (which is less of a problem) and wouldn't know how to fix it once someone inevitably reports a bug in the future (more of a problem). From my perspective, adding a feature is a commitment to users of the tool that that feature will continue to work in the future. Elastic is not the most backwards compatible thing in the world and they happily change their APIs in between versions. And even elktail's simple feature set breaks with new Elastic versions (hence multiple versions). More complex feature set has higher likelihood of breaking and higher maintenance requirements across Elastic versions.

b) Some of PRs also add superfluous changes that I don't necessarily agree with.

Issue (a) could be addressed by creating a proper integration testing environment in travis and hence building a testing foundation upon which proper integration tests for features could be built on. I'd be happy to merge PRs that have integration tests demonstrating how the feature works and thereby having a way to test something if it's reported broken.

(b) could be solved by adding contribution guidelines.

Of course, elktail is open source and anyone is free to fork it. If a maintained fork emerges I'll endorse it.

Let me know your thoughts.

[bergerx]

Thanks for the reply and providing a detailed perspective, that clarifies the case for me. Having several issues and PRs just sitting there for nearly a year led me to think if this project abandoned, and I checked the https://github.com/knes1/elktail/network to see if there is a properly maintained fork but wasn't able to see a clear leader. Seems like @SvenDowideit had an attempt and already created #30.

I think you can always ask for including some tests or comment on the ones that you don't agree with, but leaving them un-answered creates the impression that the owner doesn't care about it anymore.

Thanks for making this cool tool available for us and I appreciate your work.