Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Idea: refactor to use habitat binary #145

Open
bdangit opened this issue Apr 26, 2017 · 4 comments
Open

Idea: refactor to use habitat binary #145

bdangit opened this issue Apr 26, 2017 · 4 comments

Comments

@bdangit
Copy link
Owner

bdangit commented Apr 26, 2017

Managing binaries and install steps for each Linux distribution is a real chore. Would be nice to just manage one binary and the config. Habitat makes that possible.

Pros:

  • Less work on maintaining recipes
  • Not much effort to support future/old Linux distros

Cons:

  • consumers may not be up to speed on habitat (operational concerns)
  • installation and running is easy, just operational maintainence is not fleshed out.
@fishnix
Copy link
Collaborator

fishnix commented Apr 28, 2017

This seems cool, I'd like to see how it would work. Would we then need to manage the habitat stuff too? or does that come from upstream?

As a word of caution, it feels a little like how everyone was using runit for a while in the Chef community instead of native init systems. That almost always caused more issues than it solved and made it hard to get over the barrier of "many new/unfamiliar things on a system". (nothing against runit!) . I'm actually guilty of forking a cookbook to "just use systemd" 🤦‍♂️ because the community one installed runit.

@bdangit
Copy link
Owner Author

bdangit commented May 4, 2017

The idea would be to setup habitat on the box, run hab install core/influxdb, feed it some configs if necessary. However, I'll wait til after chefconf, to see what patterns emerge (chef + habitat).

Ya i hear you on the runit. It bit my team in the butt when we wanted to manage our own process on each host -- forgot that others also depended on that runit system and we kind of caused an outage. Hab could run in complete isolation though -- ways to do it, but may not be worth the effort.

@djoos
Copy link

djoos commented Aug 25, 2017

This is actually a really interesting point!

@bdangit: any further details on the patterns around Chef + Habitat post-Chef conference?

Thanks in advance for your feedback!

@bdangit
Copy link
Owner Author

bdangit commented Aug 29, 2017

Hi @djoos, thanks for chiming in. Have not had the chance to really commit to this idea just yet. Post conf, there's been sentiment that Chef managing Hab is more of anti-pattern. The idea is to maintain Hab at the top and it would be used to auto-magically place apps into order. IMO, it sounds like Hab is to replace/augment service supervisors like runit, systemd, etc. Then again there should be something to manage or "orchestrate" install Hab apps. However I can easily see just a bunch of bash scripts doing this function since Hab would be taking of all the automatic setup/installation/running.

With that said, I'm not against the idea of making a section dedicated to installing influxd with Hab. I think it might be better to start simple without the Chef framework to accomplish this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants