Fatal Error during import of Task and related Sub-task

[Barbara Murray]

I have created a very simple test load file for bulk loading Tasks and sub-tasks as per the Import data from a CSV instructions.

I have:

Issue ID, Parent ID, Issue Type

1,,Task

,1,Sub-task

I have mapped the Issue ID to Issue ID in the import config

and Parent ID to Parent ID in the import config

When I load I get a fatal error advising that Parent can't be NULL, but it is not NULL.

I have previously validated all of the other values  to be loaded with no errors so it is not a rogue validation error.

How can I resolve this problem please?

[Barbara Murray · Accepted solution]

Actually found this article, so the solution is load parent then load child.

https://confluence.atlassian.com/jirakb/csv-import-parent-child-mapping-1096095688.html

[Nic Brough -Adaptavist-]

Welcome to the Atlassian Community!

Your file looks technically correct to me, but I have a nagging doubt that it doesn't like the empty issue id on the sub-task line.  

When I do imports with sub-tasks, I've always done it with the issue id populated for all issues, even though I really don't need it on the sub-tasks. So my files would look like

• Issue ID, Parent ID, Issue Type
• 1,,Task
• 2,1,Sub-task
• 3,,Task
• 4,1,Sub-task

So two issues, one with two sub-tasks.

I once had a problem with the order as well - for some reason, I'd put a sub-task above the issue it was part of, and that failed to import, presumably because Jira was reading one line at a time, but that was in a very old version of Jira and I don't think it goes line by line any more, I think it reads the whole file and works out which issues need creating before their sub-tasks.

[Barbara Murray]

Hi, thanks for the response, I updated my file and loaded again but still get the same error.

Parent Issue cannot be null.

[Nic Brough -Adaptavist-]

That suggests you have a line that names a sub-task issue type, but has no parent ID filled.

One thing to try is filling in every line with a parent id - for the issue level issues, use their own id as the parent (jira should ignore the field for issue level issues)

Re-using my example with that tweak, you would import:

• Issue ID, Parent ID, Issue Type
• 1,1,Task
• 2,1,Sub-task
• 3,3,Task
• 4,1,Sub-task

[Barbara Murray]

I still get a fatal error, and these instructions do not tie in with the steps provided in the standard user instructions provided by Atlassian for importing from CSV.