Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-54383

consider adding bitbucket-oauth and github-oauth to Evergreen

    • Icon: New Feature New Feature
    • Resolution: Won't Do
    • Icon: Minor Minor
    • evergreen
    • None
    • Evergreen - Milestone 2

      Evergreen already includes the multibranch plugins for github and Bitbucket I think it is just a logical consequence to also provide the ability to do authentication via the same platforms. This will also remove the need for a "local user database".

          [JENKINS-54383] consider adding bitbucket-oauth and github-oauth to Evergreen

          R. Tyler Croy added a comment -

          I think this might be better suited for a "bitbucket" flavor batmat, that may be a way we can incorporate more plugins like this which I've never used, in a manner that doesn't bloat the other flavors.

          As an added bonus, we might be able to find somebody who actually uses bitbucket to maintain that flavor

          R. Tyler Croy added a comment - I think this might be better suited for a "bitbucket" flavor batmat , that may be a way we can incorporate more plugins like this which I've never used, in a manner that doesn't bloat the other flavors. As an added bonus, we might be able to find somebody who actually uses bitbucket to maintain that flavor

          BitBucket is actually pulled in by Blue Ocean I believe. TBH, I think we should even remove bitbucket since it's absolutely untested, and unlikely to be in the short term. I filed JENKINS-54420 to track this more global issue.

          I nevertheless agree we should find a way to propose bitbucket. I tend to think the flavor way is (as it currently stands) not an awesome way, because we are risking to see an explosion of flavors with many axis (aws-github, aws-bitbucket, docker-github,docker-bitbucket, and later with 3 or more axis can become quickly unmanageable, or when we add azure and other IaaS providers).
          That is a discussion out of scope for this JIRA though.

          Baptiste Mathus added a comment - BitBucket is actually pulled in by Blue Ocean I believe. TBH, I think we should even remove bitbucket since it's absolutely untested, and unlikely to be in the short term. I filed JENKINS-54420 to track this more global issue. I nevertheless agree we should find a way to propose bitbucket. I tend to think the flavor way is (as it currently stands) not an awesome way, because we are risking to see an explosion of flavors with many axis (aws-github, aws-bitbucket, docker-github,docker-bitbucket, and later with 3 or more axis can become quickly unmanageable, or when we add azure and other IaaS providers). That is a discussion out of scope for this JIRA though.

          while I do somehow understand your points, it sadly means I will have to wait even longer for evergreen to be useful for me

          Dominik Bartholdi added a comment - while I do somehow understand your points, it sadly means I will have to wait even longer for evergreen to be useful for me

          (Thinking out loud, mostly) In the short term, I think we could just go by adding a new flavor if that allows someone to use Evergreen for real. I think we should open gates more, even too much if needed, so that we can gather experience. (Also, at that stage, someone like you is probably even more interesting so you'd be able to understand potential issues and know better how to interact with the Jenkins community and report them.)
          Sort of allowing tech debt to accumulate temporarily to go past this stage of the product.

          But as rtyler was hinting, this probably means we'd need someone to help support that bitbucket flavor, that could be partly you imod if you agree (not asking you here for any sort of long term committment here, obviously).

          Baptiste Mathus added a comment - (Thinking out loud, mostly) In the short term, I think we could just go by adding a new flavor if that allows someone to use Evergreen for real. I think we should open gates more, even too much if needed, so that we can gather experience. (Also, at that stage, someone like you is probably even more interesting so you'd be able to understand potential issues and know better how to interact with the Jenkins community and report them.) Sort of allowing tech debt to accumulate temporarily to go past this stage of the product. But as rtyler was hinting, this probably means we'd need someone to help support that bitbucket flavor , that could be partly you imod if you agree (not asking you here for any sort of long term committment here, obviously).

          batmat that really sounds like a good way for me - I'll be more than happy to do this in some kind of supporting this special flavor. 

          For what its worth it: I currently have to move away from DEV@cloud anyway and everything I'm doing with my own instance on AWS is in the mind to be replaced by a provided solution at one day (e.g. evergreen) - this also means that I do nothing by hand anymore, everything is done with JCasC and job-dsl exclusively. So I should be able to start my seed job on any env and have the exact setup I need to do all my stuff. 

          Dominik Bartholdi added a comment - batmat that really sounds like a good way for me - I'll be more than happy to do this in some kind of supporting this special flavor.  For what its worth it: I currently have to move away from DEV@cloud anyway and everything I'm doing with my own instance on AWS is in the mind to be replaced by a provided solution at one day (e.g. evergreen) - this also means that I do nothing by hand anymore, everything is done with JCasC and job-dsl exclusively. So I should be able to start my seed job on any env and have the exact setup I need to do all my stuff. 

            Unassigned Unassigned
            imod Dominik Bartholdi
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved: