How to create a project with its history? i.e. completed sprints, etc.

[xavier le galles]

Hi,

We want to move from HP AM to Jira but unfortunately this is not straightforward. First, because there is no CSV export usable from AM. So we decided to do it by hand i.e. recreating previous sprints, main stories, etc. As the result was not conclusive, we even tried exporting/re-importing stories, a few updates in the DB, etc.

Here is our current situation and our problems:

• All burndowns are flat i.e. 0 story points 
• All sprint reports are flat (i.e. 0 story points) while the list of stories is good and the total story points displayed per sprint is also good
• The velocity report does not show any completed points but commitments are good
• The release burndown stacks all story points into the current active sprint and cannot display any forecast because "it needs at least 3 completed sprints"

I've seen a lot of threads related to one or another of these problems, so let me add a few remarks:

• All sprints have correct start/closed dates
• All stories have a status done which is well defined in the right column of the workflow.
• All stories have a created date < start date of the sprint and a resolved date < closed date of the sprint

So yes, it's a real mess! But we don't see anymore solution to our problems. Is there any documentation on how to start a project with an existing history? Is there any way to solve our problems (even by updating the db)?

Thank you

[Thomas Deiler]

Dear @xavier le galles,

for sure there is always a way - it just a matter of resources.

As you have the data already  in Jira, you need to fix the dates manually. At all, this means modifying Jira's database without Jira.

Before you do, make a backup, shut down jira and know the risk - you could destroy everything or bring the DB in an inconsistent state.

Or better, make an offline copy of the db, so you can do everything without a hurry.

Then use a tool like PL/SQL Developer to access and browse the database. Best you do this in a pair - four eyes see more.

Alternatively you can do this with a full backup of Jira - modify the XML. That depends on your abilities, where you feel more comfortable.

So long

Thomas

[xavier le galles]

Hi Thomas,

Thank you for your answer but we already tried it and did not have much success. For instance, all dates for sprints are right but stories are not taken into account: the list of stories per sprint is also right but it does cumulate their story points...

We have just an inconsistent state and I do not know where to start from. It is not a question of time/resources but more a question of documentation: there are tens of tables in this Db...

[xavier le galles]

What I mean, is that there is no obvious error. If you don't know what you're looking for, you can waste hours in this database.
If at least I could log the queries made by each report, I could understand what is wrong?

[Thomas Deiler]

Jup - the db design is a little bit complicated. And more like a forest than a tree. ;) To be honest, I have no better approach ...