Issue Details (XML | Word | Printable)

Key: JRA-7948
Type: New Feature New Feature
Status: Open Open
Priority: Major Major
Assignee: Unassigned
Reporter: Kevin James
Votes: 9
Watchers: 6
Operations

If you were logged in you would be able to see more operations.
JIRA

Ability to Create issues and automatically Clone/Move to multiple projects and maintain changes across Cloned issues

Created: 14/Sep/05 11:16 AM   Updated: 09/Apr/08 01:43 AM
Component/s: Web interface
Affects Version/s: None
Fix Version/s: None

Time Tracking:
Not Specified

Issue Links:
Part
 
Reference
 

Participants: =Neal Applebaum, Benjamin Naftzger [Atlassian], George Hickey, Jeff Turner [Atlassian], Kevin James and Tom Miller
Since last comment: 39 weeks, 4 days ago
Labels:


 Description  « Hide
This is somewhat related to JRA-2581. This is our most pressing problem regarding JIRA right now. We have a number of projects which depend on a rather large set of common libraries. When issues are reported they are reported against a given product even though the bug may be down in the common libraries affecting a number of products. Many times it is not obvious if the issue is at the application level or down in the common library when it is first reported, but the issues are always reported against the product, not the libaries. What we need is a way to easily report the same issue against multiple projects and to track them separate if need be, but also to apply changes to this issue across all the projects it is reported against.

Basically, I'd like to see a feature that allows me to clone an issue to multiple projects and versions in one step. This would be available at the Create, Edit , or Resolve issue stage. Once an issue was cloned then anytime a change was made, other than to the Version or Component, it could be applied to zero or more of the linked (of type clone) issues. This would give us the ability track issues separately if need be or to report them against multiple products and resolve them in multiple projects (for tracking purposes) easily without having to clone/move-clone/move-clone/move for each project.

Perhaps a UI where I could check a box saying to 'Apply to multiple projects' and then Add Project/Versions to a list. If it was a new issue it would automatically create the clones and move them to the specified projects and if it was an existing issue you could select which project to apply changes to and/or add/delete projects from the list.

We've tried to use the Cloning feature currently available to address this issue, but we're finding it is just to cumbersome to create, clone/move, clone/move, clone/move, etc to be practical. Also, if you apply a change to one issue such as a comment, this is not propogated to the other cloned issues. You can still get to it via the links, but it is easy for people to miss relevant information if they don't review the linked issues.



 All   Comments   Work Log   Change History      Sort Order: Ascending order - Click to sort in descending order
Benjamin Naftzger [Atlassian] added a comment - 01/Dec/05 11:53 PM
This would be nice as a workflow post-function.

The workflow transition could be called 'Copy this issue to project' or 'Assign to X Project' (for example) and the issue could then automatically be copied to the relevant project and an issue created and the status could be 'Assigned to X Project' (for example).

It would be nice to be able to configure the workflow post function to allow the user to configure the 'move' field mapping aspects (as seen in the move issue wizard) for the overall functionality of the transition so that this occurs automatically every time the workflow is transitioned.

Perhaps, to allow the post-function to potentially copy it to multiple projects the 'move wizard' field mapping process can be performed for each project. Then if more than one project exists the transition can display a project mulit-select drop down option in the transtion screen liisting the projects or display no project drop down if only a single project is targeted for the transition.


George Hickey added a comment - 09/Oct/06 10:31 PM
I have a very similar need. As changes across a number of separate units (each represented by a project) are identified, we need to raise an issue to signal action is required on most projects. This might be as high as 70 projects, meaning that we would have to clone a single issue 70 times.

We would much prefer to raise an issue across multiple projects or to raise the issue on one project and then easily copy it to many others.

Has any simple way been identified to do this? Alternatively, is any work underway to include it?


Jeff Turner [Atlassian] added a comment - 10/Oct/06 11:57 PM
This talk of cloning issues and then trying to keep them in synch suggests something is broken in the data model. There should be only one location to record info.

If the problem can be summarized as:

> When issues are reported they are reported against a given product even though the bug may be down in the common libraries affecting a number of products.

Wouldn't the best solution be to distinguish between "product" bugs and "library" bugs?

  • A product bug is reported from the user's point of view, in terms of end-user functionality (eg. "mail doesn't send").
  • A library bug is a technical description of what went wrong in a library (eg. "deadlock in mail queue library").

They could then be linked together (ie. product bug "is caused by" library bug).


=Neal Applebaum added a comment - 11/Oct/06 09:03 AM
Jeff - here's my use case. We have opened up JIRA to our customers for use as a helpdesk. They have access to one (support) project only. Most of the issues can be resolved by fixing data, doing training etc. However, sometimes, they find a real bug that needs to be fixed. So we close the support issue and create a new issue in our internal development project, linking the two together. I've actually created a workflow action called "Close for Development to Fix" which
  • automatically sets the resolution to "Tracked in Development"
  • automatically creates a new issue (clone) using a plugin hacked up between me and another user
  • sets security of new issue so client cannot see it
  • links the 2 issues so the support issue says "Tracked in development issue XXX-123" and the development issue says "Derived from support issue YYY-456"

The customer doesn't see the link (except in change history, if they poke around), but internally we do. So we can track support issues to development issues (and fix versions) and vice versa.

Think about how you link support requests on support.atlassian.com to bugs on jira.atlassian.com. It's a different instance (which people which only one license can't do, by the way). But what's more there's no internal link between the 2 issues. Not the best model.

However, the summary of this issue states:

maintain changes across Cloned issues

Now that is not something I need (or want), and I think that's the part that you are commenting on. So if that's the crux of this issue, I may change my vote. I'm looking for a clone/move operation - both as part of a workflow operation and standard operation. By the way - I've tried Mei's plug-in, but it doesn't work for me because:

  • it does not respect permissions
  • it doesn't work in Weblogic
  • the newer doesn't work at all for me (works until JIRA is restarted)

Jeff Turner [Atlassian] added a comment - 12/Oct/06 12:03 AM
> maintain changes across Cloned issues

Now that is not something I need (or want), and I think that's the part that you are commenting on.

Yes. It's fairly clear that a support case is distinct from a bug, and shouldn't be kept synchronised. I am suggesting that a similar distinction could be made between product bugs (end-user-oriented) and library bugs (programmer-oriented). If such a distinction cannot be made, then there shouldn't be two separate projects.

> So if that's the crux of this issue, I may change my vote. I'm looking for a clone/move operation - both as part of a workflow operation and standard operation.

Clone-and-move would be handy. I think it should be its own issue.


Tom Miller added a comment - 08/Jan/08 01:05 PM
We actually have a separate project called CORE to handle the common library issues. No need to duplicate across products.