How to add, populate and sync a custom field to a running project?

Question

We were asked to create a JIRA project based on a MSAccess database. We now have a JIRA project that contains around 2.5k Issues and each issue contains around 120 fields whose contents were originally imported via a CSV file.

The 120 required fields were the result of months of preparation and planning with the customer - paring down the Access database's original 180 fields.

In the months that followed, the customer worked in the JIRA project - manually updating some of the Issue fields and has creating new issues.

A few days ago the customer decided they should have included a fields that was not in the original requirements and we were asked to add a new custom field - which we did => Issues->CutomFields->Create.  However, the customer was very suprised when they discovered there was no data in any of these new custom fields for any Issue.

The customer had expected that by adding new field (which IS in the Access db) it would somehow be populated with field data we did not originally import.

Question: Is there a way to get data into this new field - which must pair up with all of the previously imported data for each Issue?  I really don't see how -unless we can somehow stipulate that for Issue JIM-00001 NewField=x, for Issue JIM-00002 NewField=y for Issue JIM-00003 NewField=z, etc.

Any ideas?

unknown · Answer

Andrew,

I finally updated the fields.  BTW: I'm using 7.10.0 JIRA Software - Server

Using just the Issue Key and field I wanted to update would not work because JIRA required the Summary field.

NOTE: I did have to

* create a new CSV import file with Issue Key and Summary data plus the new field and updated values
* export Issue Key and Summary data from JIRA as a CSV
* merge the 2 and
* use the merged CSV for the import.

I mapped

* Issue Key->Issue key  (in the System Fields dropdown list below Custom Fields)
* Summary->Summary (in the System Fields dropdown list below Custom Fields)
* Field2UpDate->Field-to-Update (a custom Date field)

and it worked.

This was the 3rd import attempt. The 1st 2 created duplicate Issues - so I must have done something wrong.

ALWAYS

* create a backup before any work so you can do a System Restore if things don't work out as planned
* use the JIRA Administration->System->External System Import->CSV for all import work

Jim

unknown · Answer

Hi Andrew,

I was actually thinking of this but couldn't imagine it working that easily.

Based on your answer, I'll give it a try via my testing platform.

Thanks for your reply.

Jim

Andy Heinzer · Answer

You can do a CSV import again into Jira.  Provided you setup your CSV file to have at least two specific columns:

* One with the issue key in Jira for that issue (ex ABC-123)
* The other column/field will need to contain the new field values that correspond to that issue.

But considering that the old system clearly doesn't have Jira issue keys, that can be difficult to match this up.   If you have some other field value that is unique to each issue in the old system that also still exists in the new system, then you could use that to match up these values.

Essentially, you might have to export all the issues from both system into two different CSV files and then merge these into a single CSV file that contains both the Jira issue key and the new field record.    When you perform the External CSV import in Jira, and you map the issue key, if that issue already exists, then the CSV import can be used to simply update existing issues rather than create new issues.

Thinking about a bit further, you might also need to export from Jira (into CSV), the description, and summary fields from your issues so that you can reimport those same values.   I have this itching feeling that the CSV importer won't let you continue unless you define those fields in your import file, even if your intention is only to update existing records.   Before trying this for all the records, I would recommend just trying this for one test record to make sure you know exactly which fields are needed to make this work.

But even this method would only work for issues that existed in the old system.  The new issues in Jira won't have any data/value for this field yet since they didn't exist in the old system.