Celebrate Excellence in Education: Nominate Outstanding Educators by April 15!
Found this content helpful? Log in or sign up to leave a like!
I'm trying to add the 'integration_id' fields to a batch of users (just working on Canvas TEST instance at the moment) but I'm getting the above error message when I come up with a csv file with 4 columns only:
integration_id, user_id, login_id and status
(user_id, login_id and status are required; integration_id is what I'd like to add)
I am getting a warning for *each* user in the file .These are all existing users, so there's no question about them being duplicates. I'm simply trying to add a value to the 'integration_id' field in each of the users.
Any suggestions? (I've seen a similar post from a few years ago but I don't think the response applies, since I'm getting this for every single record in that csv).
Solved! Go to Solution.
Hi @pgo586,
Try this format... Leave new_id blank (since it's not being changed). Also ensure you have the type column and for each line the value of that column is "user" (since this file can also update courses or other objects too). I just tried this in my test instance and it added an integration_id to an existing user with the sis_id given.
-Chris
Hi @pgo586,
I think you may need to use the change_sis_id.csv file format instead of just trying to add the information in the user file to add that info to existing users. Could you try that and let us know if it's successful?
-Chris
Oh, interesting @chriscas , I didn't even realize that that file even existed actually!. Of course I'm not trying to change the sis id values at all, but I could try it out and see if it works though. Thanks for suggesting that!
Before I had a chance to test this, I realize that I had made a silly mistake: I hadn't named my csv file 'users' in the first place :-(. When I did this, the SIS import went through. UPDATE: this didn't work after all (even though the import went through, the 'integration_id' values were not filled). See next reply.
Sorry for replying above too quickly. Turns out that, even though the SIS import had gone through, the integration_id fields did not get set unfortunately So I went ahead and tried your suggestion @chriscas , but this time around I got another error (officially a warning, but it really tells me it didn't work): "A new_id, 'blah', referenced an existing user and the user with sis_user_id 'blah' was not updated". So it looks like the suggestion didn't work, after all. Any further advise? Thanks in advance.
for more details, I used the following columns: new_integration_id, old_integration_id, new_id, old_id and 'type' where:
- the old_integration_id values are empty
- the new_id and old_id values are the same (since we're not changing sis ids, just adding integration_id)
- the type is 'user' for all the records
Hi @pgo586,
Try this format... Leave new_id blank (since it's not being changed). Also ensure you have the type column and for each line the value of that column is "user" (since this file can also update courses or other objects too). I just tried this in my test instance and it added an integration_id to an existing user with the sis_id given.
-Chris
That actually worked this time around, and apparently the field values were set as desired. Thanks so much @chriscas!
To participate in the Instructure Community, you need to sign up or log in:
Sign In