How do the JOBS work in IAS?
A Read job checks only for changes in the source system. If there have been changes in the target system, they are not affected by the job.
How to delete inactive users with the IAS Jobs. Maintain only the active users
How to ensure that inactive users are continually deleted?
To keep the IAS database with only the active employees, it’s necessary to configure the source system filters to only read the “active” users. So, when an employee is deactivated or deleted in SAP SuccessFactors, the user will be deleted with the next Read Job.
When the job runs, re-write all the data base, so if the user becomes “inactive” then it wil be deteled.
The source system filter has to be configure with sf.user.filter = status eq ‘active’. to ignore the inactive and to delete with the “read job”
Ensure that your job run is free of errors. If the particular error does not allow IPS to properly calculate the entities that are no longer present in the source system, the delete operation will be skipped.
Although the name of the job implies reading users, a Read job could also lead to deleting users from the target system. This could happen if any of the following changes occur after you have run a provisioning job (Read or Resync):
- A user has been deleted in the source system. As a result, on the next run of the Read job, the deleted user won’t be read from the source system and will be deleted in the target system.
- A user filter is applied in the source system. As a result, on the next run of the Read job, the users not matching the filtering criteria will be deleted in the target system.
- A condition is applied either in the source or in the target system. As a result, on the next run of the Read job, the users not matching the condition criteria will be skipped or deleted in the target system.
- The mapping of the targetVariable entityIdSourceSystem or entityIdTargetSystem has been changed.
How to schedule a Resync Job?
To run a scheduled sync between your source and target systems, you need to set up the following source in the system property: ips.full.read.force.count.
For example, ips.full.read.force.count=10 results in alternating full reads after every 10 delta reads are performed. This property only impacts scheduled runs; manually triggered runs are ignored.