- Product: Alma
How to delete user records based on a subset of users (e.g. students whose expiry date has already passed)?
The utility job "Purge Users" should be used.
Click here for the Online Help page regarding "Purging Users".
Click here regarding "Configuring Delete User Policy Settings".
To note, below is an outline of a process which enables more control and caution over the exact list of users to be purged.
The steps are:
1. In Analytics, create a report for the users to be purged/deleted. Include the Primary ID, for easier next steps.
2. Export the results to an Excel spreadsheet.
3. Keep only the Primary ID column, and at the header (Cell 1, Col. 1) put: USERNAME. See a related Knowledge Article, "Create User Set based on Analytics".
4. As precaution, add or ensure that the list of users all belong to an unmistaken, distinct User Group, For example, call that User Group: "Purge_Students_2017". Add in the following 2 tables:
User Configuration Menu > “User Groups”
User Configuration Menu > “User Record Type/User Group”
* Note: the advantage is one more check, or better control, over the users to be purged. For Analytics and statistical purposes, please consider a meaningful name for the user group.
5. Create an Itemized Users Set, in: ADMINISTRATION > Manage Jobs And Sets > Manage Sets.
- Use the above Excel spreadsheet to ensure exact group to be purged/deleted.
6. Update the Expiry Date and User Group for this set, by navigating to: Run a Job > Type: Users > Update/Notify Users.
- In this job, select Purge Date for a date in the past (e.g. a week ago).
- Update the User Group for the users, e.g. "Purge", as defined in step 4 above.
7. Finally, Purge users by going to: ADMINISTRATION > User Management > Purge User Records.
- Number of Days after Purge Date can be blank (more in the Online Help, link given below)
- Specify correct User Record Type, e.g. "Public", otherwise the job will ignore these users!
- Specify the above unique User Group, e.g. "Purge".
- "Waive Threshold" - our experience lead us to recommend using the value of "1" or higher.