Actually, I disagree with Al on this one.
You do not need to repeat the cohort column if you are adding the user to the same cohort each time. You only need to add the user to the same cohort once. That's what this column does: it makes the user
a member of the indicated cohort. Once a member, there is no point in trying to add them again.
If you add more than one column with the same cohort name for a user, the
upload process will just ignore those additional ones since the user is already a member of the cohort after it processed column cohort1. It will not toss an
error, it just ignores the request for adding the user again to the same cohort.
Consider this example:
username,firstname,lastname,email,cohort1,cohort2,cohort3
test,Test,User,test@example.com,CohortA,CohortA,CohortA
which attempts to add the user to the same cohort three separate times, which is not necessary. It will add the user to CohortA when it processes column cohort1 and it will ignore cohort2 and cohort3 columns since it checks to see the user is
already a cohort member when it tries to add them to a cohort.
All you need is this:
username,firstname,lastname,email,cohort1
test,Test,User,test@example.com,CohortA
Or just this if you are only doing cohorts:
username,cohort1
test,CohortA
All the cohort column does is add the user as a member of the indicated cohort. It does not do any enrolments direclty: that is handled by the Cohort enrol sync process as usual when adding users manually, and it in turn handles any enrolments into any
courses that the cohort is connected to an an enrolment method.