...
Following values are not stored currently and needs to be fixed:
- x-app-id header is currently being read as X-app-id. As such, this value is currently being sent as null. This needs to be fixed. We are planning to store this in 'pdata.id'.
- x-device-id header is currently NOT being read. This needs to be fixed. We are planning to store this in 'did'.
2) AUDIT events are required to be generated for all objects - For ex: User, Course, CourseBatch, Badge etc
...
3) Object or actor types should be in upper camel case. For ex: User, Content, Resource, Collection, Group, UserLocation etc. Distinct object types are listed in this sheet https://docs.google.com/spreadsheets/d/1se19XYQf7gxb3ZuPQL-ZcmWTn2utIKiBdNg2XMXMXAc/edit?usp=sharing and should one of the entry in the above sheet. (Reach out to me if any new type needs to be added)
Following object types need to be added.
- BadgeClass
- Location
- SystemSetting
- Note
- Role
- Dashboard
We are currently using Batch. Should we change this to CourseBatch or the excel sheet can be updated?
Code changes are need to ensure object types are in upper camel cases. At some places object type is currently being sent in lower case.
...
Working as expected (i.e. size set is the total documents matched as per given filter criteria and not the total documents returned in response as per limit).
c) edata.topn should contain the top n (default configuration as 5) object ids returned by the search
...
7) Correlation data needs to be passed wherever applicable in all audit events. For ex: If a user is enrolled into a course batch and an AUDIT event is generated with object.id as UserEnrollment, CourseId and CourseBatchId should be added as “cdata” for the same.
Should type in cdata be upper camel case?
Can we use 'batch' or do we need to specify cdata.type as 'CourseBatchId'?
Can we use 'user' instead of 'UserEnrollment'?
8) Each batch created for sync should not have more than 200 events
...