If you are encountering repeated rollup failures. There are several steps you can do to research this.
If the rollup is failing on a “master data not yet indexed” error, then
- 1) Check for any failed attribute change runs. If there are any, these need to be corrected first.
- 2) Check the application log for the master data daemon indexing job.
- Run T-code SLG1 with:
- Object: RSDDTREX
- Subobject: TAGGRFILL
- Ext. Identifier: *NEWMD_INDEX*
b. This will display all logs for the day, scroll down to the bottom for the most recent.
c. If you are seeing a lot of red statuses in the last 10-20 logs, then check the errors within them
- If the error is a lock on an info object. This means that most likely a load is currently running into this object. The lock should clear itself within 30 minutes as the load completes. If it does not clear, it is most likely due to attribute change or load that has failed and is not completed successfully.
ii. If the error is a BWA bad return code, then you have some indexing issue and if this error repeats across more than 10 of the most recent logs, then you must drop the indexes on the cube, and then recreate them.