Troubleshooting Live Data Connections to On-Premise Data Sources

This lesson is designed as an overview of some of the tests and tools that are commonly used to troubleshoot live data connections to on-premise data sources.

As SAP Analytics Cloud administrator, you will be able to run some of the tests yourself using the SAP Analytics Cloud performance tools, however for some tests, you will have to work with your SAP system administrator or network administrator to solve the connection issues.

There are some important tools available in SAP Analytics Cloud that help you to track performance and optimize your system performance.

In this video, we walk you through some of the key performance tracking functionality in SAP Analytics Cloud.

The overall performance of SAP Analytics Cloud is mainly determined by the three components:

1. Client time (frontend)
2. Network time (between the frontend and the backend)
3. SAP Analytics Cloud service time (backend)

The Measurement Tool helps you to benchmark your client hardware and measure the client latency and the network bandwidth to the SAP Analytics Cloud service.

You can choose two options:

1. Run Client Test
2. Run Network Test

The Run Client Test tool runs several scripts on the local machine and calculates a normalized score. This score is generated based on best practices for system performance for SAP Analytics Cloud.

To run the test:

• From the Measurement Tool, select Run Client Test.
• When the test is complete, the System Measurement Score is displayed.
• Click the score to review information on the System Measurement Score that has been provided.

Tests with different CPU generations showed that SAP Analytics Cloud performance is CPU sensitive. To improve your client score while keeping your existing hardware configuration, watch out for CPU-intensive applications and processes that can have a negative impact on your SAP Analytics Cloud performance.

Some examples of these applications and processes include:

• Business video conferencing services
• Virus scanning
• Synchronizing of shared network drives

The Run Network Test measures the client latency and bandwidth to the SAP Analytics Cloud Service.

1. From the Measurement Tool, select Run Network Test. The scores are displayed once the test is complete.
2. Network Latency: A calculation based on your latency and system bandwidth. The latency corresponds to the elapsed time between the request and the response to the SAP Analytics Cloud server. Latency, measured in milliseconds (ms), describes the time it takes from starting the request until the response arrives. A good latency is a ping rate within a low milliseconds range.
3. Download Bandwidth: The amount of data transferred from the SAP Analytics Cloud server to the client. It is measured in megabits per second (Mbps).
4. Upload Bandwidth: The amount of data transferred from the client to the SAP Analytics Cloud server. It is measured in megabits per second (Mbps).

The network scores can be negatively affected by other applications consuming the bandwidth, various network configurations, or different network issues.

Some examples of these applications include:

• Internet server provider connection
• Virtual Private Networks (VPNs)
• Firewalls

As there are quite a few ways to track performance in SAP Analytics Cloud, we encourage you to explore the tools that are available to you as an SAP Analytics Cloud administrator.

Take some time now to visit the SAP Help Portal to investigate the specific tracking tools available to you.

• Analyze System Performance | SAP Help Portal
• Optimize System Performance with the Analysis Tool | SAP Help Portal

A detailed list of troubleshooting topics can be found in the 2511489.

SAP Analytic Cloud leverages SAP's proprietary Information Access (InA) infrastructure for live data connection to various sources. InA enables applications to perform analytics, planning, and search the data that is stored in SAP systems. The InA model allows for the definition of queries using rich or complex semantics. Generally speaking, InA is a definition for a JSON object model that reflects the analytic or planning request of the user.

In order for live connections from SAP Analytics Cloud to succeed, the InA server must be correctly deployed in the SAP system to which you are connecting.

In this lesson, we'll use SAP HANA as our on-premise source system to demonstrate how to verify the InA server. In the example below, you can see:

1. The lock icon is displayed in the browser address bar, indicating that it is protected by SSL. The browser should not throw any SSL certificate validation error. Not secure indicates an invalid SSL setup.

   The test should be done using Fully Qualified Host Name (FQDN) of the SAP HANA server. Valid SSL certificates are always tied to FQDN, so the IP address or short name of server should not be used.

2. SAP HANA’s GetServerInfo URL.

   SAP HANA GetServerInfo: https://<xs-host:port>/sap/bc/ina/service/v2/GetServerInfo

3. The JSON response.

A valid JSON response should be shown as in the preceding example. An error in the JSON response, as shown below, indicates that the SAP HANA system doesn’t have the required EPMMDS package installed.

The process to test the InA service for SAP NetWeaver systems such as SAP Business Warehouse, SAP S/4HANA, and SAP Business Planning and Consolidation is the same as the one described in this lesson.

SAP NetWeaver Systems GetServerInfo: https://<NW FQDN>:<NW Port>/sap/bw/ina/GetServerInfo

While the process to test InA for SAP Universes is the same, the SAP Live Data Connect component must be deployed before running the InA service validation test.

Live Data Connect GetServerInfo: https://<LDC FQDN>:<LDC Port>/sap/boc/ina/GetServerInfo

In this lesson, we will use Chrome as our default browser, however, it is possible to use developer tools in Edge as well.

Before you start, in SAP Analytics Cloud, create a new live connection or edit an existing connection. Do not hit OK to create/save the connection. It is at this point that you will launch the Chrome developer tools and capture the HTTP trace.

1. Launch the developer tools from your browser or by pressing CRTL+SHIFT+I .
2. Go to the Network tab and make sure that the circle in the top-left corner is red, if it is gray, click it to start recording. You can also clear the log by clicking the circle with the line through it.
3. Make sure Preserve log and Disable cache are enabled.
4. Return to SAP Analytics Cloud and select OK button in the connection.

During the live connection attempt, several GetServerInfo calls are made. To analyze the information that was captured, in the Network tab, choose the first GetServerInfo call and review the Response Headers.

You should see the CORS headers issued from the backend systems (e.g., SAP HANA, SAP BW). If the CORS headers are not present, then the connection attempt has not been successful and is an indication that the CORS setup in the backend system needs to be reviewed.

The logs found in the Console tab of the debugger shows the full description of any errors encountered during the HTTP(s) calls. Console logs are useful to troubleshoot potential connectivity issues to data sources.

The capture records the time it took for each HTTP call. For live connections, you can monitor the GetResponse HTTP calls to understand how long it takes for the call to process. If you see abnormally high time, it could indicate potential network bottleneck or the slowness at the database level.

If you are using SAML authentication with SAP Analytics Cloud, it's helpful to be able to capture and decode the SAML assertion when troubleshooting.

The easiest way to do this is to add the SAML Chrome Panel extension to the Chrome developer tools. This tool can be added from the Chrome Web Store.

You can use the SAML tab to capture and decode the SAML assertion. This capture is useful to troubleshoot:

• User authentication issues to SAP Analytics Cloud. In such scenarios, you can capture the SAML assertion and analyze value for the NameID to confirm it matches exactly what's specified in SAP Analytics Cloud. For example, if you expose email as the NameID, then it must match exactly (it is case sensitive) to the email of the user in the SAP Analytics Cloud system.
• SAML SSO authentication issues for live connections to data sources such as SAP HANA.

In the following example, you can see a capture for a login to the SAP Analytics Cloud system used for this course.

1. The SAML tab in the Chrome developer tools.
2. The NameID attribute in the SAML assertion.
3. Additional attributes custom1, custom2, custom3 attributes that are used for team and role assignment.