Celebrate Excellence in Education: Nominate Outstanding Educators by April 15!
Found this content helpful? Log in or sign up to leave a like!
Hello,
With the deployment of the 1.3.0 client, I've decided to TRY and migrate from Postgres into my native SQL Server environment.
I have successfully initialized ~88 tables, but I know that's not enough. I'm seeing an issue within the Assignments table that probably exists in other tables.
The script that's creating the Assignments table isn't creating the column "submission_types" and manually adding the column doesn't have any effect.
I have reviewed the CREATE TABLE script generated by the client and that column definitely isn't there. The reference table is being created, but the constraint is not.
Has anyone else successfully deployed this into SQL Server? Has anyone else found similar issues?
I do have a ticket into Support on this.
Thanks,
Jim Sorenson
Hey @JamesSorenson!
I'm sorry to read this. Can you please help me by providing additional details about this?
Feel free to send me these in private message if you feel like.
I sent you a private message...tried to reply twice and it never posted.
@JamesSorenson I'm just back from holidays, I have got your message. I will get back to you ASAP.
I've also been having this exact issue with assignments and submission_types. Keeps getting hung up there and failing.
EDIT:
Wondering if this is a versioning issue? We are using our Azure SQL Server, which shows as version 'Microsoft SQL Azure (RTM)' and not SQL Server 2019.
For the record, this is the same root cause as mentioned in a later thread: submission_types in the table assignments is an array of an enumeration type but Microsoft SQL Server supports neither arrays nor enumeration types. Normally, an array would expand into a separate table with a foreign key linking back to the original table. However, an enumeration also needs to expand into a separate table for database engines that don't support enumeration types out of the box. This leads to a many-to-many relation, which is a more complex table (consisting of three columns: unique ID, foreign key to table assignments, and foreign key to table of enumeration values) that is, in turn, more challenging to populate. For this special case, one possible solution is to store enumerations as their string value in a JSON array embedded in a column of type varchar. This solution is currently being tested, and will likely be rolled out shortly.
I didn't receive a notice of resolution on my ticket, but this appears to have been fixed as of last week?
I've been running daily manual updates using WSL and a standalone Linux system and assignments had been failing, but miraculously assignments initialized last week.
Can you confirm this is now resolved?
Yes, it has been resolved, I can confirm. Sorry for not updating you!
To participate in the Instructure Community, you need to sign up or log in:
Sign In