After upgrading jira from 9.4.14 to 10.3.2. everything is working except the bulk import.

[testcase1062]

hi 

After jira data center version upgraded from 9.4.14 to 10.3.2. we successfully upgraded and everything is fine and working I didn't check the bulk import that time. Now, when ever we are importing the bulk issues in new project after carefully mapped all the fields and run it shows the error in import users section and below picture appears. If I click next the issue keys automatically taken as a users and create that issue key in jira internal directory while we are using the LDAP.

Any solution and suggestions please

[David Nickell]

If you are dealing with a relatively small universe of users, I recommend using the "Map Field Values"  Checkbox to convert "Bad Input" to "Valid User".    Sure -- it's a little scary having to use any workarounds.... but it is probably the fastest path.   Test a couple of records before going "all in" on every issue

[Noah Leu]

Hello @testcase1062 ,

does this error apply to all issues you have imported or is this a problem of some specific issues from the import file?

It sounds like there may be a mistake within the import file that is causing the issue keys to be used as users. If you select the "Map field values" checkboxes on issue key and user fields, do you see any issue? 

In the case the mapped values show some unusual behaviour you may want to check if the separator is correctly select or the "Email Suffix for New Users" is correctly set up. 

Can you provide some more input to the structure of your import file as well as the import configuration? Be aware to not publish any secure information or company data. 

Best regards
Noah

[testcase1062]

Not all issues, only specific ones, are affected.

My CSV file format includes the following columns: Summary, Issue Type, Issue ID, Issue Key, Child Link, Parent ID, and Epic Link.

In the hierarchy, Epic and ISLM Milestone are at the same level. When an Epic is created, ISLM Milestones are mapped within its issues. Now, the ISLM Milestone is linked to the Epic, serves as the parent of Task issue types, and Tasks further contain Sub-tasks.

Issue hierarchy:
Epic ← ISLM Milestone ← Task ← Sub-task

The issue occurs  the ISLM Milestone issue is treated as a new user, and Jira automatically creates it in the internal directory with the following format:

Username: Xyz-1

Email: Xyz@noreply.com

This unintended user creation happens during the import process.

[David Nickell]

Ah.. you may be searching for a stray comma or other value that is getting your CSV fields adjusted by accident.   You may be able to change the order of the fields in the CSV.    please take note too -- when you try to fix stuff like this in Excel, I've seen it display one thing to me (like a date format) but actually masking the real way it is stored in the text version of the CSV.    

[testcase1062]

I tested the same CSV file in UAT environment all good and fine. But in the prod it behaves like this.

Should I do full reindex. It might works or not?

[testcase1062]

My problem is resolved thanks for the help. @David Nickell @Noah Leu

While mapping the parent link, in my instance there are two parent link fields are there, In that one is user picker, mistakenly I'm taking the that user picker field. That's why it creates the issue key as a new users.