How to start a sprint in a very large project using (rapid) boards?

[Ex Employee Aaron Bockelie]

I am having difficulties wrapping my head around starting a sprint when there are a number of different components of the team:

Each component lead controls a different set of issues. They are differentiated primarily by labels, but sometimes by component and labels, or other criteria.

These set of issues provide the basis for the board filter.

Each lead starts their respective sprint, based on a team-wide schedule - currently, two week sprints.

The problem:

Occasionally there will be issues that are already involved in a sprint someone else has started. This precludes the actual owner from starting their own respective sprint.

Attempted workaround:

I attempted to make an uber board that includes every issue in the project, and start a sprint with that board. This halfway worked - some boards respected that a sprint had started and the board user could drop issues from backlog into that sprint. However, other users sometimes did not see the sprint that was started, even though their issues matched the filter from uber board. I know this is not the correct way to do it, but I expected behavior that was consistent.

My question is this: how do we manage multiple sprints where issues are stored between sprints? Apparently greenhopper allows an issue to be in one sprint and not a different one.

[boardtc]

The main problem seems to be delineating your stories/issues so there is not such an overlap between teams, ideally there should be none and this should be the goal. There should be no reason you can't adapt your issues/stories/sub-tasks to cope with this.