Issue Summary

In "Issue Edit screen" the options for 'Select List (cascading)' being defaulted to (not available) which is not being recognized by Jira, neither the value was inserted manually.

Steps to Reproduce

1. Create a custom field of type 'Select List (cascading)' and give the options for the select list, this happens only for the second level value in cascading list.
2. Go to filed configuration and make the field as "required" that we just created.
3. Create a new issue where give the parent value of cascading list an option, and secondary value leave as 'None' and click on "create".
4. Once the issue gets created click on "Edit" button on the left side and try to update this will throw an error stating "The filed is required" and the secondary field where we left the option "None" will be defaulted to "(Not available)" which you can't see while creating an issue and this value is being defaulted only in the "Edit issue" screen.

Attached screen recording while reproducing the issue and this seems to be related to the bug JRASERVER-62455 where it was introduced to prevent the data loss. and this was introduced in Jira version 8.8.1

Expected Results

When the field value is set to "None" this should not default to "(not available)" in the Edit screen.

Actual Results

While clicking on the "Edit" option the values which have "None" as default being changed to (Not available), make Jira treat the not-empty field as empty. And issue editing is only possible when the value is manually changed to None. This thing is the main problem.

Workaround

You can edit a velocity template for the cascading select custom field to fix this issue.
To do so:

1. open the file <JIRA_INSTALLATION>/atlassian-jira/WEB-INF/classes/templates/plugins/fields/edit/edit-cascadingselect.vm.
2. There, edit the code on line 48 from:

   •         #if ($databaseParent && !$hadSelectedChildOption)
     

   • to:

             #if ($databaseChild && !$hadSelectedChildOption)
     

3. After applying the change, save the file and restart the Jira node.

In Jira DC, the workaround needs to be applied separately for every node.