In this document we will discuss how to report an API (OData) performance issue to Cloud Product Support (CPS).
In the rare occasion that you experience High Performance levels running your OData integrations (e.g. an UPSERT or GET), the below information is essential (in terms of a thorough investigation) and will help CPS deliver the root cause of the issue.
- SuccessFactors (SF)
Reproducing the Issue
- View your OData API Audit Log.
Log into your SF instance and navigate to the OData API Audit Log (via Admin Center). You have noticed a High Performance time on a query or upsert:
1. Enable your OData API Audit Log Availability:
Admin Center --> Tool Search --> API Center --> Click on "Audit Log Setting":
Click"Enable all Payloads" option this will enable audit logs for all operations (read/Write) and for (Success/Failed) status.
Payloads will be available only for 4 hours.
Reference Guide: 2639894 - [1805 Feature] API Audit Log Availability
2. Gather the OData Log information:
Navigate once again to the OData API Audit Log section and view the API Operation that caused the High Performance level.
Please provide the below required information (if possible):
The HTTP Message (all body and header information):
NOTE: Please provide the HTTP Message information in a .txt file (non-photo/screenshot format) and attach them to the incident.
This is crucial as they contain unique information.
Below is a sample photo of were you can gather this information:
Reference Guide here: 2209703 - How to check and understand Odata API Audit Logs
If possible, please provide a timeframe comparison (accompanied by time and date) of High Performance rates vs Low Performance rates.
Example: On the 28th JAN the response from an UPSERT of EmpJob was 14.771ms. But, on 29th JAN the same UPSERT response was 98.13ms
An OData API Refresh can resolve some API Performance issues. Detail on this here: 2171588 - OData API Metadata Refresh and Export - SuccessFactors