Problem:
During second import the uids are added instead of overwritten.
You have an import file as *csv and on first import you get a user called i.e. “paulb1”. Running the second import this user not recognized by the importer and creates a new user called “paulb2” despite of the same record-uid and source-uid.
Environment
Make sure the ucr variable ucsschool/import/generate/user/attributes/no-overwrite-by-schema is not set to an empty string.
Unset the variable or set it to the desired values. When unset it defaults to “mailPrimaryAddress uid”. The default prevents the importer from overwriting these two attributes when creating a new user.
When set to an empty string the importer generates an uid and does not overwrite it from the existing user- thus creating a new one.
UCRV help:
Whitespace separated list of LDAP attributes that the user import should not overwrite if their values are non-empty in LDAP, empty in the input data (CSV) and a schema for them exists. Default value if UCR variable is unset: “mailPrimaryAddress username”
Solution
Set the variable either to your desired values or unset it properly to use the defaul:
ucr unset ucsschool/import/generate/user/attributes/no-overwrite-by-schema
See also: