JIRA Importing/Updating subtasks - csv

Question

Im somewhat new to JIRA (skill level novice)
Jira v 6.4.8
JIM v 7.0.12
I am attempting to import issues using the Issue->Import from CSV (bulk create tool)
I have a defined ticket CM-1 as a parent ticket. A generic CSV looks like this
Summary, Parent ID, Issue ID
CM-2, CM-1,
CM-3, CM-1,
CM-4, CM-1,
The first import works successfully and maps as children to CM-1
We attempt to re-import (to update the ~100 fields that changed overnight, not shown in this example for clarity)
Summary, Parent ID, Issue ID
CM-2, CM-1, CM-2
CM-3, CM-1, CM-3
CM-4, CM-1, CM-4
We encounter an issue where new subtasks are created, and nothing is updated.
I have also tried to map the Issue ID found when a I inspect the subtask tickets XML. It looks something like this
<item>
<title>[CM-2] CM2</title>
<link>
https://website.net/browse/CM-2
</link>
<project id="11902" key="CM">Change Management</project>
<description>CM-2 Description</description>
<environment/>
<key id="191147">CM-2</key>
<summary>CM-2</summary>
Specifically the ""
So that would look like
Summary, Parent ID, Issue ID
CM-2, CM-1, 191147
CM-3, CM-1, 191148
CM-4, CM-1, 191149
Once again we see new issues created and no updates performed. I've read the documentation, searched their 'Answer's' asked multiple questions, searched everywhere, but im not seeing any solutions. We literally need to update thousands of tickets, at least once a day - we don't have the manpower to perform this task any other way.
Criteria:
This needs to be able to be performed by an end user or a team lead, they will have access to the bulk import tool (Bulk create) from the Issues-Import issue From CSV link but will not have access to the administrator level external project imports.
I know this isn't an ideal long term solution, and would like to investigate a method to further automate this but we need a solution short term (this).
I appreciate any and all responses. We are importing from a very outdated instance of remedy that's going to remain in use for the next ~3+ years.
Thanks,
Jacob

Answer 1

First of all, if you want to update issues via CSV, you must include an 'Issue Key' column and, during import, map it to the issue key field (CM-1,CM-2 etc. are issue keys in your example). Otherwise every import will generate new issues in JIRA.
The 'Issue ID' and 'Parent ID' columns refer to internal IDs (not issue keys). For adding/updating sub-tasks, you need to figure out the ID of the parent (see below), and in the CSV, write the parent id in the 'parent ID' column, and leave the 'issue ID' value empty. This is explained in the 'Creating sub-tasks' section here.
Figuring out the id of an existing JIRA issue is somewhat tricky (unless you import them from the beginning with your own internal ID which has some sense). An easy way from the GUI is to right click the Edit button and choose 'open in new tab'. Then, the URL of the edit page will include the id (e.g. http://jira-srv/secure/EditIssue!default.jspa?id=91796).
If you need to automate it, you will have to resort to directly querying the database (unless someone else can offer you a better way... as far as i know the REST API does not expose it). See the discussion here if you want details.