Details

    • Epic Name:
      Orchestration Improvements
    • Epic Status:
      To Do

      Description

      Goal is to separate concerns between orchestration of services (what to launch or stop at which point in time) and internal scheduling (offer matching, state handling of instances, Mesos interaction).

      The current deployment abstraction is very opinionated towards rolling upgrades, doesn't scale well and is not very flexible when it comes to integration with other components like load balancers or user specific upgrade or failure handling strategies.

      This was previously covered under Actions or Improved Deployments (MARATHON-7005).

      Other issues to consider while scoping:

        Attachments

          Issue Links

            Activity

              People

              • Assignee:
                Unassigned
                Reporter:
                thomas Thomas Rampelberg (Inactive)
                Team:
                Orchestration Team
                Watchers:
                Ananth cb, Arthur Rand, Chris Gaun, Chris Lambert (Inactive), daltonmatos, drewrobb, Judith Malnick (Inactive), Lee Munroe (Inactive), Marco Monaco, Matthias Eichstedt, Victoria Saunders
              • Watchers:
                11 Start watching this issue

                Dates

                • Created:
                  Updated: