Symptom
You have an issue with SAP Analytics Cloud (SAC), Analytics Designer (AD) and want to know how to troubleshoot it.
Image/data in this KBA is from SAP internal systems, sample data, or demo systems. Any resemblance to real data is purely coincidental.
Environment
SAP Analytics Cloud, Analytics Designer
Resolution
When facing an issue with SAP Analytics Cloud, Analytics Designer, please check the following:
- What is the model type that the analytic application (AA) is based on?
An import model (from BW, BW/4HANA, HANA, Excel spreadsheet, etc.) or a live model (BW, BW/4HANA, HANA, UNX, etc.)? - Does the same issue or behavior occur within a standalone story (built in the story designer)?
If the issue happens with a standalone story, then the issue should be reported to the team responsible for the model type used in the story: - BW or BW/4HANA live model → LOD-ANA-LDC-BW.
- HANA live model → LOD-ANA-LDC-HAN.
- Import model → LOD-ANA-DES.
- Planning specific workflow → LOD-ANA-PL.
After these checks, we can take a closer look into two common issues:
1. Performance:
For performance troubleshooting, we analyze two things:
1. Backend response time:
- Run the analytic application, press F12 (or go to the browser menu (...) > More Tools > Developer Tools).
- There are two types of requests: GetServerInfo and GetResponse. For performance purposes, the focus is the GetResponse request.
- Click the GetResponse record and go to "Preview".
- Check the run-time.
- If the run time is high, the issue may have to be processed by the team responsible for the model or connection type (import model / live connection).
- Make sure other potential causes have been explored and eliminated such as network latency.
2. Check the widgets and scripts that take the most start-up time:
- Launch the application.
- Add the following parameter in the URL: "?APP_PERFORMANCE_LOGGING=true".
- Open the browser Developer Tools (F12) and paste the following on the Console tab: window.sap.raptr.getEntriesByMarker("(Application)").filter(e => e.entryType === 'measure').sort((a,b) => (a.startTime + a.duration) - (b.startTime + b.duration)).
- This will generate data regarding the Application performance as follows:
3. After checking where the time is spent most, refer to the performance hints guidelines, check the script and settings based on existing best practices below:
2. Scripting Errors/ "How to" Requests:
1. Since Analytics Designer parses JavaScript code directly, one way to analyze scripting errors is by setting several "console.log" API’s into the application.
We can then follow the variables in the browser developer tools console tab. In the example below, there is a for loop (line 17) within a for loop (line 13):