Data Warehouse / Analytics Cannot Distinguish Code Or UDF Mappings Between Multiple P6 EPPM Data Sources In A Single Data Warehouse Schema
(Doc ID 2428980.1)
Last updated on APRIL 13, 2022
Applies to:Primavera Analytics - Version 18.104.22.168 and later
Primavera Data Warehouse - Version 22.214.171.124 and later
Information in this document applies to any platform.
When Data Warehouse is holding multiple datasources for P6 EPPM, W_XLAT_S (the translation table holding code/udf mappings) will get updated to reflect the translations for the associated datasource when the ETL is executed.
This makes W_XLAT_S incapable of distinguishing mappings (udf or code) between multiple P6 EPPM schemas. This also results in Analytics from reflecting proper translation-to-mapping IDs.
Example using project code mappings:
- DATASOURCE 1 maps:
- CodeA to ID 1
- CodeB to ID 2
- CodeC to ID 3
- DATASOURCE 2 maps:
- CodeD to ID 1
- CodeE to ID 2
- CodeF to ID 3
- When running the ETL for datasource 1, the translation table is updated to reflect datasource 1 translations.
- When running the ETL for datasource 2, the translation table is updated to reflect datasource 3 translations, causing incorrect translation names to display for datasource 1 mapping.
This will create issue/confusion when creating analyses since the incorrect translation name is displayed.
The issue gets compounded when the translations are being mapped to non sequential values because now it will display a single list of translations for all P6 EPPM datasources, where the translation name will update for the datasource ID an ETL is executed against but leave the non-mapped ID values the same (since they relate to other P6 EPPM datasources). This can cause duplicated names/translations to display (if a code/udf share names between environments but map to different IDs), or display mappings that dont relate between datasources. See Project Activity And UDF Codes Missing Or Write To Wrong ID In Star Schema After ETL Run (Doc ID 2430317.1).
The translations for code/udfs must properly distinguish between multiple P6 EPPM environments.
STEPS TO REPRODUCE
- Setup Data Warehouse for two P6 EPPM datasources
- For datasource 1, map 3 codes/udf (to ID 1,2,3)
- For datasource 2, map 3 codes/udf (to ID 1,2,3) with different names
- Execute ETL for datasource 1
- Execute ETL for datasource 2
- Note the translation names for datasource 1 are no longer valid in W_XLAT_S.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document