Show TOC

Function documentationRoot Cause Analysis and Exception Management

 

Root Cause Analysis

Today's distributed, multi-technology customer solutions offer multi-channel access through diverse devices and client applications. Analyzing the root cause of an incident in such systems requires a systematic, top-down approach. This approach must be guided by tools that help customers to perform this analysis as efficiently as possible, following SAP best practices.

The objective of Root Cause Analysis is to isolate the problem and provide:

  • Immediate corrective action (workaround) to restore service operations as quickly as possible with minimal disruption to end users

  • Complete solution to the issue at hand

Root Cause Analysis tools are designed to reduce the number of resources required in each step of the resolution process. In addition, the Root Cause Analysis infrastructure is open to the integration of new SAP technologies, applications, and third-party software.

Example Example

An end user experiences slow performance on a local browser. The performance problem may be caused by the client, the network, or some component of the server environment, which itself is comprised of different instances of different technologies. Server-side processing may take place in an SAP Enterprise Portal (based on Java technology), then reach out to an SAP ECC back end (based on ABAP technology), and finally call the database and the storage subsystem for data retrieval. A performance problem or functional defect may occur at any one or several stages of this round trip launched by the end user from the browser. The Root Cause Analysis tools in SAP Solution Manager help to identify the specific component that caused the bottleneck.

End of the example.
Exception Management

Exception Management enables you to centrally monitor and handle business-critical exceptions in your system landscape. In the Exception Management Cockpit, you can correlate, analyze and process exceptions, and use predefined guided procedures to handle errors quickly. Exception Management supports monitoring of:

  • Technical, single exceptions for any system type managed by SAP Solution Manager. You can also monitor exceptions that occur in supported cloud services, including SAP Cloud for Customer, SAP HANA Cloud Integration, SAP SuccessFactors, and Dell Boomi, in the context of an SAP-based hybrid scenario.

  • Process-flow-driven, multiple-step exceptions that occur within interfaces and processes that are distributed among multiple components in an ABAP system landscape that can also include third-party systems. To enable logging of technical and business exceptions during runtime, you use SAP Exception Management Instrumentation Platform (SAP EM-IPA) to mark critical locations in ABAP source code.

Exception management collects single exceptions and multiple-step exceptions in a central exceptions store. The exceptions store also provides exception data to the central Alert Inbox and other SAP Solution Manager monitoring applications based on the End-to-End Monitoring and Alerting infrastructure (MAI).

Prerequisites

  • For information about the prerequisites for Root Cause Analysis, see SAP Note 1483508Information published on SAP site.

  • Administration has configured the managed systems to include in your analysis in SAP Solution Manager Configuration.

  • Administration has set up service connections to SAP so that you can get support for the managed systems in your solutions. For more information, see Service Connections.

Features

End-to-End Workload Analysis

If a customer experiences a performance issue, E2E Workload Analysis might be the tool to start with. SAP Solution Manager regularly collects performance data for your managed systems and makes this data centrally available in the Root Cause Analysis work center. The Workload Analysis tool allows you to identify general performance bottlenecks caused by sizing problems, or performance problems that affect all users of a particular system.

Workload Analysis

End-to-End Trace Analysis

End-to-End Trace is a tool for isolating a single user interaction through a complete landscape, and providing trace information on each of the involved components for a single interaction only, starting with the user interaction in the browser and ending with data being committed to the database. The most common use case for E2E Trace Analysis is to pinpoint the cause of long-running user requests within a complex system landscape, but you can also identify which functional errors have occurred at any point during the execution of a single request. These exceptions (such as a dump) are then attached to the trace. The trace can then be used to do functional testing and to ensure that an activity that is executed in one system does not lead to functional errors in connected systems.

Trace Analysis

End-to-End Change Analysis

If a system behaves differently after a certain date or change, E2E Change Analysis is the first tool to use. It displays changes (such as transports, support package updates, and profile parameter changes) that have been applied to a system within a certain time frame. The Change Analysis application is based on a central Configuration and Change Database (CCDB).

The CCDB provides the foundation for change analysis and change reporting based on daily configuration of system, host and database related configuration data. A change analysis can then be performed on all information stored in the CCDB. You can also compare different systems and generate a report. This approach identifies the problem by comparison with another system rather than by drilling down, which is faster and easier in most cases.

Change Analysis

End-to-End Exception Analysis

E2E Exception Analysis provides unified access to exceptions reflecting in high severity log entries and dumps. Exception Analysis provides the basis for statistical analysis on exceptions in the landscape, but also allows you to access component-specific log and dump viewers directly, by navigating to the appropriate tool on the managed system.

Exception Analysis

System Analysis

For Java based systems, the Introscope transaction trace can be used to identify which part of a request in a Java environment caused the problem. Introscope is shipped by SAP with preconfigured dash boards, offering dedicated views for the SAP Application Server Java. A deep analysis of Java problems is possible in the investigator mode, which displays detailed performance metrics. Several other system analysis tools are provided, including Thread Dump Analysis, Change Reporting and Central Log Viewer.

System Analysis

Host Analysis

With the Host Analysis section of the RCA Work Center it is possible to analyze the most important OS metrics like CPU, memory, paging, network and disk/file system. In addition the filesystem browser allows a central read-only access to predefined directories on the managed system without having to logon to the console of the managed system. In the same manner it is also possible with the OS Command Console, to execute predefined, non destructible commands (like a ping, netstat, iostat and so on). Both tools work to ensure that even whilst access to the managed system is granted, no change can be made, no business data can be accessed and not harmful commands can be executed.

Host Analysis

Database Analysis

The DB Analysis summarizes DB Performance Warehouse specific metrics across all supported DB types. This also includes standalone DBs that are connected to the DBA Cockpit within Solution Manager

Database Analysis