Closed JoshVanL closed 2 years ago
/assign @munnerz
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: 7ing, JoshVanL
The full list of commands accepted by this bot can be found here.
The pull request process is described here
Can we add an e2e test that:
1) creates a pod that uses a CSI volume 2) asserts a CR exists 3) deletes the pod 4) asserts the CR does not exist
Otherwise, lgtm :)
@munnerz yes, sounds good to me!
The project doesn't have any e2e dep setup yet, so will do that in a separate PR containing the boilerplate. How do we feel about experimenting with nix for managing the deps? :slightly_smiling_face:
I am of the opinion that nix flakes is growing in popularity and will become a "default" dependency management and distribution tool. csi-lib is a good project to experiment with given it has minimal end-user facing interaction and there is not much (basically only the current Dockerfile) to convert.
/lgtm
@munnerz sorry, wasn’t clear and should have added a hold 😅 I meant let’s hold this PR and rebase on top of the e2e test boilerplate once that had merged. No biggie, I’ll create an issue.
@munnerz yes, sounds good to me!
The project doesn't have any e2e dep setup yet, so will do that in a separate PR containing the boilerplate. How do we feel about experimenting with nix for managing the deps? 🙂
I am of the opinion that nix flakes is growing in popularity and will become a "default" dependency management and distribution tool. csi-lib is a good project to experiment with given it has minimal end-user facing interaction and there is not much (basically only the current Dockerfile) to convert.
I'm impartial, though given the extra complexity & knowledge required to manage/operate a nix based setup (of which as a project I don't think we have much experience), I'm wary to introduce a whole new ecosystem (especially as, in this case, we already have prior art for setting up dependencies).
This is a good place to experiment given the limited developer community around it, and minimal size. That said, for me at least there are a lot of unknown unknowns, which makes me wary to give a firm "let's do it 😄".
@munnerz sorry, wasn’t clear and should have added a hold 😅 I meant let’s hold this PR and rebase on top of the e2e test boilerplate once that had merged. No biggie, I’ll create an issue.
😄 woops! But I think this is a valuable change for us to get merged anyway, especially as this is in the csi-lib (and we can leave it untagged if we aren't yet confident in its results) and is a really narrow change. An issue would be great 🥳
Fixes #29