SAP Business Objects GRC 10.0 Automated Monitoring...

SAP COMMUNITY NETWORK SDN - sdn.sap.com | BPX - bpx.sap.com | BOC - boc.sap.com

© 2011 SAP AG

Applies to:

SAP® Business Objects™ Governance, Risk and Compliance (GRC) Process Control release 10.0. For

more information, visit the Governance, Risk, and Compliance homepage.

Summary

Automated monitoring of business process controls is a key feature of SAP Business Objects Process

Control (PC) 10.0, and a fast-evolving feature category in the market. This document presents an overview

of the monitoring features of PC 10.0, which will help put more detailed user guides in context.

Author: Atul Sudhalkar

Company: Governance, Risk, and Compliance

SAP BusinessObjects Division

Created on: 15 August 2011

Version 1.6

SAP Business Objects Process

Control 10.0 Automated

Monitoring Overview

https://www.sdn.sap.com/irj/sdn/bpx-grc


© 2011 SAP AG

Typographic Conventions

Type Style Description

Example Text Words or characters quoted

from the screen. These

include field names, screen

titles, pushbuttons labels,

menu names, menu paths,

and menu options.

Cross-references to other

documentation

Example text Emphasized words or

phrases in body text, graphic

titles, and table titles

Example text File and directory names and

their paths, messages,

names of variables and

parameters, source text, and

names of installation,

upgrade and database tools.

Example text User entry texts. These are

words or characters that you

enter in the system exactly as

they appear in the

documentation.

<Example

text>

Variable user entry. Angle

brackets indicate that you

replace these words and

characters with appropriate

entries to make entries in the

system.

EXAMPLE TEXT Keys on the keyboard, for

example, F2 or ENTER.

Icons

Icon Description

Caution

Note or Important

Example

Recommendation or Tip


© 2011 SAP AG

Table of Contents

1. Business Scenario............................................................................................................... 1

1.1 The SAP Business Objects GRC Solution ................................................................... 1

1.2 Automated Monitoring .................................................................................................. 1

1.3 Usage: Planning and Testing ....................................................................................... 2

1.4 Usage: Monitoring......................................................................................................... 2

1.5 Audience ....................................................................................................................... 3

2. Background Information ..................................................................................................... 3

2.1 Automated Monitoring Overview .................................................................................. 3

2.2 Queries and Events ...................................................................................................... 4

2.3 Data Sources ................................................................................................................ 5

2.4 Business Rules ............................................................................................................. 5

2.5 Architecture ................................................................................................................... 6

2.6 The Business Context .................................................................................................. 6

2.7 Planner ......................................................................................................................... 7

2.8 Scheduler ...................................................................................................................... 7

2.9 Event Monitor ................................................................................................................ 7

2.10 Legacy Automated Monitoring ...................................................................................... 7

3. Prerequisites ........................................................................................................................ 8

3.1 Systems Installation and Activation .............................................................................. 8

3.2 Post-installation Configurations .................................................................................... 8

3.3 Master -Data Preparation ........................................................................................... 14

4. Monitoring Methods .......................................................................................................... 15

4.1 Data Sources and Business Rules ............................................................................. 15

4.1.1 Design-time .................................................................................................... 15

4.1.2 Runtime .......................................................................................................... 32

4.2 Data Source and Rule Types ..................................................................................... 34

4.2.1 SAP Query Data Sources and Rules ............................................................. 34

4.2.2 Configurable Data Sources and Rules .......................................................... 36

4.2.3 BW Query Data Sources and Rules .............................................................. 45

4.2.4 ABAP Report Data Sources and Rules ......................................................... 46

4.2.5 Netweaver Process Integrator-based Data Sources and Rules .................... 46

4.2.6 Web Services-based Data Sources and Rules.............................................. 47

4.2.7 SoD Data Sources and Rules ........................................................................ 47

4.2.8 Event-driven Data Sources and Rules ........................................................... 48

4.2.9 ABAP Programmed Data Sources and Rules ............................................... 48

5. Further Reading ................................................................................................................. 48

6. Appendices ........................................................................................................................ 50

6.1 Appendix A - Configurable Rule Example .................................................................. 50

6.2 Appendix B - Change Analysis Example .................................................................... 54

6.3 Appendix C: Data Source and Rule Type Summary .................................................. 63

7. Acknowledgements ........................................................................................................... 66


© 2011 SAP AG

8. Copyright .............................................................................................................................. 1

SAP Business Objects GRC 10.0 Automated Monitoring Overview

December 2011 1

1. Business Scenario

1.1 The SAP Business Objects GRC Solution

Governance, Risk and Compliance Management covers many enterprise-wide perspectives, including

regulatory compliance, risk measurement and mitigation, monitoring to ensure good governance, and

other similar concerns. The assumption is that the basic business processes necessary for running

any business—purchasing, sales, hiring and promotion, etc.—are already in place. SAP

BusinessObjects Governance, Risk and Compliance (GRC) solutions provide an overview of such

processes from a risk and compliance point of view, and help customers measure risks and monitor

compliance.

1.2 Automated Monitoring

This document focuses on one specific aspect of GRC: automated monitoring of backend systems

and processes, part of the Process Control 10.0 application (PC 10). Customers of GRC use

automated monitoring for configurations, master data and transactions; SAP BusinessObjects Process

Control 10.0 (PC 10) provides a range of techniques to address these needs. While particularly well-

suited to monitoring SAP‘s backend applications, PC provides a sound platform for monitoring other

applications as well. Such other applications include ERP and related software suites from other

vendors, but can also include IT management, physical access management, software used for

tracking movement of goods and managing logistics, and so on.

SAP partners also provide products which integrate with SAP GRC products to provide joint solutions

for diverse needs. Many such solutions are showcased on the SAP Ecohub website, and many of

these include automated monitoring with PC.

The following figure depicts how GRC fits into the corporate IT landscape, and into a corporate

governance and compliance strategy.


December 2011 2

Note that, in this view, monitoring is front and center in any effective GRC practice. For example, a

customer can monitor configurations of SAP Sales & Distribution (SD) credit checks to make sure that

even as these are changed during routine maintenance, they remain compliant with company policies

(see the appendix in section ‎6.2 for details). PC 10 automated monitoring capabilities enable such

monitoring of configurations and master data, and also backend transactions. The possible solutions

shown at the top of the picture include content which might involve joint solutions with other SAP

products, partnerships with other software vendors, or a partnership with consulting partners who

provide domain expertise by industry, region, or line-of-business practice.

1.3 Usage: Planning and Testing

In PC 10.0, the Planner helps control owners and testers test the effectiveness of existing controls,

and plan compliance activities over a specified period. Tests of effectiveness can be manual, using

well-understood procedures, but these can also be semi- or fully automated tests. The techniques

described in this document can also be used to perform some controls, and can help control owners

gather information relevant to performing their jobs.

1.4 Usage: Monitoring

Beyond the Auditors‘ perspective in testing controls, there are many situations in which customers

want to ensure that business processes meet their expectations. As explained in this document, PC


December 2011 3

enables customers to capture their expectations of how these controls should be configured, and how

transactions should occur. Correct configurations ensure that business process steps that those

configurations control will always comply with customers‘ intentions; broader transaction monitoring

can then be used to cover those situations where configuration-based controls are not enough, or to

look for fraud at the margins.

In a sense, automated (or semi-automated) monitoring can also help individuals perform the control

function. For instance, a person responsible for reviewing and approving purchases might want to

look at background information on the requester, vendor, pricing trends, etc. before making a decision.

Workflow can route the requisition itself to his or her inbox, but PC automated monitoring can provide

the additional information needed to actually reach good decisions.

1.5 Audience

This document aims to introduce customers and partners to automated monitoring in PC 10.0. The

topic is somewhat technical, so at times this document introduces language and concepts which might

challenge the reader.

In GRC, we call process experts, compliance and risk officers, audit professionals and senior

managers and executives ―business users‖. We use ―functional experts‖ to mean professionals who

know a particular business process well, including knowing the business process steps and

transactions, configurations and master data. The term ―technical experts‖ refers to software

professionals who understand databases, queries, web service configurations, or programming.

―Implementation experts‖ are professionals who know the PC product well—they will be responsible

for installing and configuring it, or upgrading from previous releases. Finally, ―rule designers‖ will be

implementation experts with particular expertise in configuring monitoring rules.

This is an introductory document, which we hope will be useful to all three groups of people. It should

give business users a feel for what sort of monitoring is possible, without requiring them to understand

every section. Functional experts should be able to understand most of this document, and technical

experts will no doubt want more details than we provide here. Section ‎5, Further Reading, gives links

to more detailed documents.

2. Background Information

2.1 Automated Monitoring Overview

To monitor any system in your IT landscape, PC first has to be able to extract data from it. The data

could be anything: configurations, master data, transactions, usage logs, or any structured information

which the monitored system can provide on demand (or which it can proactively send to PC—see

section ‎2.2, Queries and Events). GRC must know the location of the system, how to reach it (the

communication protocol), how to authenticate itself to the remote system (login credentials), where in

the system the data of interest resides, and so on.


December 2011 4

The previous figure graphically depicts the complete process for setting up and using automated

monitoring in PC 10.0. The remainder of this document will focus on only the first step: defining Data

Sources and Business Rules. Along the way, though, features are called out which help tie these

rules into the overall context in which they run, and for that reason you may find it helpful to refer back

to this figure at various points along the way.

2.2 Queries and Events

The monitoring methods available to PC customers fall into one of two broad classes: query-driven or

event-driven. PC initiates query-driven monitoring, typically via the scheduler. This is why some

practitioners also call it schedule-driven monitoring. The common characteristic of these monitoring

methods is that the monitored system is passive—all action is initiated from the PC side. The data

might come from a query, a report, a function invocation, or from any other technical source, but the

semantics are those of a query: PC invokes a particular source (query or otherwise), passes context

parameters to it, and receives back data of a known schema.


December 2011 5

Event-driven monitoring, by contrast, is not initiated by PC. An external system decides when

something is significant enough to be communicated to PC, and initiates data transfer by raising an

event. PC treats such events as data sources much the same as a query-driven data source, and

makes the event details available to business rules for further evaluation. Partners have explored this

interface to help customers monitor and remediate problems such as inappropriate network traffic or

unusual usage patterns in application logs.

2.3 Data Sources

PC can pull data from remote backend systems by multiple mechanisms. To keep track of these, rule

designers create objects called Data Sources, which store the information about the actual sources of

data on remote systems which they will invoke when a monitoring rule runs.

Data sources come in many variations (described below), to match the many different types of

information customers might want to extract, from different types of monitored systems. But as PC

Data Sources, they also present a uniform look-and-feel to the rest of PC, so that the variability of the

different types is hidden from the higher levels of the monitoring function.

Section ‎4.2, Data Source and Rule Types, describes all the types of data sources supported in PC

10.0 and their characteristics, and briefly discusses the situations in which each would be suitable.

2.4 Business Rules

Once PC pulls in data from backend systems being monitored, it applies user-specified logic to decide

if there are any problem situations (which PC calls deficiencies). In PC, such user-specified

monitoring logic is stored in Business Rules.

Business rule definitions can be simple (―A one-time vendor should not be used more than once in a

quarter‖) or involve additional calculations, grouping and aggregation (―Total spend on a one-time

vendor must never exceed $10,000‖). Even more complex logic, or specialized user-defined functions

can also be configured in the rule engine.

Section ‎4.2, Data Source and Rule Types, describes how rule characteristics vary depending on data

source types, and the implications of each.


December 2011 6

2.5 Architecture

This document gives a high-level overview of architecture for business users of SAP BusinessObjects

Process Control 10.0.

Section ‎2.1 gave an overview of the end-to-end process of defining monitoring rules and using them in

PC 10.0. But most of this document focuses on the first step: defining data sources and business

rules. That is because the first step is unique to automated monitoring, while the remaining steps are

more about all the other PC functionality.

The following figure is a conceptual diagram of the architecture.

Monitored systems are backend applications such as SAP ERP, CRM, etc. For legal reasons, this

document uses only SAP applications in examples of monitored systems, although PC 10.0 can be--

and is–used to monitor a wide selection of non-SAP backend applications.

Data sources are objects in PC 10.0 which tell PC how to extract data from backend systems being

monitored.

Business rules encode the actual monitoring logic the rule designer wants. A business rule is

designed to work against one data source (although one data source can serve more than one

business rule). That‘s because the rule engine (in which rule designers express their monitoring

intent) needs to know which fields are available for building the rule, and that depends on the data

source being used.

The purpose of a Data Source object in GRC is to know how to talk to an actual data source (such as

a query) in a remote system, and provide the data to associated rules. In ABAP, the standard

mechanism for doing this is by means of an RFC destination , which is defined in the SM59

transaction. There are nine different types of sub-scenarios which have dedicated data sources

supported in PC 10.0, described later in this document.

2.6 The Business Context

Customers use PC 10.0 to perform and test controls, and to monitor configurations and master data

settings. In this document, the term ―monitoring‖ labels all such activities. In PC master data


December 2011 7

structures, such controls are defined in the context of specific business processes and organizational

structures

To enable this, controls are monitored by assigning rules to the local instances of controls. The

business process and organizational unit associations of local controls provide the business context.

PC allows customers to use these environment variables, historically called Organization-Level

System Parameters (OLSP). The idea is that monitoring rules should not assign explicit values to

parameters in the query for company codes, plant codes, purchasing organizations, sales

organizations and the testing period. Instead, PC binds values to most of these parameters from the

context, or OLSP. Test period end dates come from the scheduler.

2.7 Planner

Use of monitoring rules via the planner is similar to their usage through the scheduler, with one

difference: the scheduler allows repeated execution of monitoring rules on a schedule, while the

planner permits only a single execution.

2.8 Scheduler

Monitoring rules can also be used for configuration, master data and transaction monitoring. For

controls which reside in backend systems, PC monitoring can ensure that the control‘s configuration is

correct; for transactions which lack a built-in control, PC can catch improper transactions.

The scheduler allows users to fire rules (strictly, all rules assigned to a control in the context of a

business process, org unit and regulation) on a specified schedule and frequency.

2.9 Event Monitor

The PC Event Monitor is the complement of the Scheduler. As is explained in section ‎4.2.8, Event-

driven Data Sources and Rules, not all data sources are triggered on a PC-specified schedule. PC

provides an event interface, by which external systems can decide when a transaction or combination

happens which needs to be presented to PC for monitoring. In such cases, PC only needs to wait for

such events to be communicated to it via an event interface.

The purpose of the Event Monitor is to allow users to instruct PC which events they want PC to be

prepared to receive via the Event Monitor.

2.10 Legacy Automated Monitoring

The previous release, PC 3.0, had a monitoring engine. More than 150 monitoring rules were

delivered with the product, many of them GRC ABAP reports (see section ‎4.2.9, ABAP Programmed

Data Sources and Rules) deployed on backend ECC systems to be monitored. Many customers

customized these delivered rules, or created their own.

PC 10.0 represents an advance in monitoring capabilities in response to customer demands. To

ensure continuity of functionality for PC 3.0 customers who upgrade to 10.0, and to protect their

investment in any custom rules, PC 10.0 includes a compatibility module which runs PC 3.0 rules

without modification. Any rules present at the time of upgrade from PC 3.0 (SAP-delivered,

customized or customer defined) will continue to work as assigned to controls, and as scheduled.

Furthermore, customers can use the new features in PC 10.0 automated monitoring for creating new

rules even as they continue to use old PC 3.0 rules through the Legacy Automated Monitoring rules

compatibility module. One limitation here is that legacy rules and rules in the new PC 10.0 framework

cannot be assigned to the same control.


December 2011 8

3. Prerequisites

3.1 Systems Installation and Activation

The PC 10.0 installation guide available on SAP Service Marketplace gives details about installation

and configuration of PC 10.0. The rest of this section addresses configurations unique to automated

monitoring.

3.2 Post-installation Configurations

To monitor backend systems, most of the customization work happens in connector configuration.

The following figure depicts the steps needed.

Creating RFC destinations (called ―connectors‖ in GRC) is standard Netweaver functionality,

accessed via transaction code SM59. With such connectors, you then configure PC to know

which connectors it should use for automated monitoring. The following figure shows the

transaction SPRO in the PC system.


December 2011 9

Use the path Governance, Risk and Compliance Common Component Settings

Integration Framework. The first of the links in the highlighted box, Create Connectors, is a

shortcut to SM59 for creating or maintaining connectors.

The next link, Maintain Connectors and Connection Types, takes you to the following screen.


December 2011 10

The three highlighted connector types are of interest in automated monitoring. Local system

connectors are used to integrate with the SAP BusinessObjects Access Control application for

monitoring segregation-of-duty violations (see section ‎4.2.7). Web service connectors are used

for external partner data sources (see section ‎4.2.6). SAP system connectors are used in all

other cases.

The next step is to define which of the connectors previously defined in SM59 can be used in

monitoring.

In the system used to capture the screenshot examples, SMEA5_100 is a connector to an ECC

system. Note in particular the third column that lists the name of a connector which is defined in

the monitored system, and which is configured to point back to the GRC system being

configured here. That is, in the highlighted row, SMEA5_100 is a connector in the GRC system,

and it points to an ERP system which is to be monitored. SM2 is a connector on the (remote)

ECC system, which points back at this GRC system. The GRC plug-in in the monitored system

uses this ―source connector‖ to call back with results of monitoring in some situations. For the


December 2011 11

most part, PC users need not be concerned with when such callbacks need to happen (see the

Technical Settings subsection in section ‎4.1.1.2), but it is necessary to complete the

configuration step.

Next examine the Define Connector Group screen, as shown in the following figure.

No configuration should be needed here, but it is shown for context. All the connector

configurations for automated monitoring should belong to the configuration group called

Automated Monitoring (shown highlighted).

Return to the Define Connectors screen, and the system connector used for your monitored

system. In the example, that connector is SMEA5_100, which points to an ECC system.

Now choose the link Assign Connectors to Connector Groups.

assign the connector in question to the ―AM‖ connector group.

Next choose Maintain Connection Settings, as shown in the following figure.


December 2011 12

A screen displays, asking which Integration Scenario you want.

Choose ―AM‖ for automated monitoring. the following page displays.


December 2011 13

The highlighted box shows nine entries called ―sub-scenarios‖—these are different types of data

sources and business rules supported in PC 10.0, and covered in more detail in section ‎4.2,

Data Source and Rule Types. to create a specific data source type (say, ―configurable‖) for a

system to be monitored, the corresponding connector must be ―linked‖ to that sub-scenario.

Select the sub-scenario you want (―configurable‖ in the example), and then choose Scenario-

Connector Link in the left-hand panel, as follows.

the following screen displays.


December 2011 14

If the connector you want to use for that scenario is not already in the list for that sub-scenario,

choose New Entries to add it. We recommend the following pattern for convenience.

System Type Sub-scenarios to enable for system

connector

ABAP applications such as ERP,

CRM, etc.

ABAP Report, SAP Query, Configurable,

Programmed

SAP Business Warehouse BW Query

Non-SAP system which is web

services enabled

External Partner (also known as web services,

or GL-MQT)

Netweaver Process Integrator (for

connecting to other systems)

PI

Same GRC system SoD Integration

Note that the Event sub-scenario is not in the table. GRC receives events via a special web

services interface (different from the one used by the web services data source type above).

This is handled in more detail in section ‎4.2.8.

3.3 Master -Data Preparation

Before monitoring rules can be scheduled to run, they must be hooked up to the regulations, controls,

and business processes, which are master data for PC.

NOTE: Setting this information up is a prerequisite for using automated monitoring, but it is not

exclusively related to automated monitoring. For more information, see PC documentation (links in

section ‎5, Further Reading).


December 2011 15

4. Monitoring Methods

This section explains the overall process of defining monitoring rules, and implementing them against

backend systems. It discusses in detail the differences between the many supported techniques (sub-

scenarios), and explains when to use them.

The goal of this document is to give only an overview of automated monitoring in PC 10.0, so it omits

the details of design-time activities. This document focuses on explaining the features rather than the

details of the design itself.

4.1 Data Sources and Business Rules

Section ‎1.2 includes a graphic of the overall monitoring process. This section describes in much more

detail the first step in that figure: defining data sources and business rules.

Data Sources in PC 10.0 encapsulate many different ways PC can extract data out of monitored

systems, while still presenting a uniform interface to rule designers who want to filter and manipulate

the data they extract. Business Rules hold the processing logic for such filters, calculations and the

logic to determine if any extracted data represents a problem which control owners need to review or

remedy.

In PC 10.0, many of the key monitoring capabilities come from the wider range of data source types

available to users, and the increased power and sophistication of those types previously available in

PC 3.0. Most of the other advances lie in the flexibility and power of the rule engine used to define

Business Rules. As said previously, the rule engine in PC 10.0 is the Netweaver BRF+ engine, with a

specialized user interface (UI) for routine usage in PC.

4.1.1 Design-time

The overall process of creating data sources and business rules is the same for all the varieties

described in further detail in section ‎4.2. There are, of course, variations but the overall process has a

strong underlying theme.

All design-time user interfaces are located under ―Rule Setup‖ in the top-level toolbar, as highlighted in

the following figure.

The Rule Setup user interface may contain many sections, depending on your role and how it is

configured in your system. The following figure shows only the Continuous Monitoring section, which

may be one of several sections visible to you.


December 2011 16

4.1.1.1 Creating Data Sources

Choose Data Sources in the previous picture. The Data Sources screen displays. The screen lists the

Data Sources previously configured in this system. You can create a new data source by choosing

the Create pushbutton.

A Data Source is defined across three tabs, of which the first two are significant to the functionality

itself. The third tab is purely for documentation and links to outside systems.

Name and Description: The Data Source name should be something descriptive which will help you

to find the data source, and help document its purpose. The description can expand on the name, but

we suggest that the name itself should be constructed to be as useful as possible. Note that like many

data types in Process Control and Risk Management applications, the name itself is not a key field in


December 2011 17

the database sense, and PC does not enforce uniqueness of names. The key for most of these

master data types is a generated number, which is seen in the first figure of section ‎4.1.1.1 as the

column titled Object ID.

Validity Dates: Validity dates determine the range of dates over which data sources, rules, controls,

and so on, can be put to use in monitoring. The ―Valid From‖ field defaults to today‘s date, and the

―Valid To‖ date defaults to infinity (that is,12/31/9999). Validity dates for data sources and rules

interact with other date ranges: for controls, business process definitions, test plans, etc. PC

monitoring works only where these date ranges all overlap. If there is no specific reason to fix the

date range, it is wisest to default the valid-from date to the start of the year, and set the valid-to date to

the year 9999.

Status: Data sources start with the status ―New‖. You can change most attributes of the data source

while it is in this status, but you cannot use it to support rule creation or any other downstream activity.

From ―New‖, a data source can be changed only to ―In Review‖; after review, it can become ―Active‖,

which is the state in which it can be used to create monitoring rules.

Search Terms: These are tags which can help in finding the right data sources, for instance when you

want to update or edit a data source, or you want to find one to reference when creating business

rules. The list of tags available is set up during customizing. You can assign up to five tags for most

master data elements, but they can be chosen only from the drop-down list, which in turn only includes

terms created during customizing.

Use The Object Field tab to define more functionally relevant attributes of the data source.

The ―Sub Scenario‖ dropdown list shows nine options; these are the different types of data sources

available in PC. These are described further in section ‎4.2, Data Source and Rule Types.

From an end-user perspective, the job of a Data Source in GRC is to provide a business-user-friendly

view of technical data sources (such as tables, fields, views, queries, reports, etc.) in monitored

systems. PC helps you do this by allowing you to replace the default descriptions with more

descriptive text. For instance, the following figure shows the vendor master table LFA1 of SAP ERP.


December 2011 18

The highlighted column shown in the following figure is editable, allowing the designer to replace the

default text with something better suited.

Connectors

For most sub-scenarios, you must define a ―main‖ connector that points to the backend system

against which PC will try to validate your definition. The only exceptions are the SoD Integration (see

section ‎4.2.7) and Event (see section ‎4.2.8) sub-scenarios. Note that after you have defined the data

source, you can use that definition to monitor other systems (that is the purpose of additional

connectors). In fact, we recommend that customers create

The following figure shows the Attachments and Links tab.

4.1.1.2 Creating Business Rules

Business rules filter the data stream coming from data sources, and apply user-configured conditions

and calculations against that data to determine if there is a problem which requires attention. In PC

this is called a deficiency.

The nature of the business rule depends strongly on the data source type, which is why the process of

creating a business rule begins with data source selection, as shown in the following figure.


December 2011 19

Unfortunately, this makes it difficult to give a general overview of business rule creation. The following

screenshot shows the full range of power in a business rule, but this range is available only with a few

data source types. Section ‎4.2, Data Source and Rule Types, describes the individual data source

types and their corresponding business rule types.

Basic Information

The name, description, validity dates, status and search terms fields serve exactly the same function

as the corresponding fields in data sources, which were described in ‎4.1.1.1, Creating Data Sources.

The Category and Analysis Type fields are dependent on the data source type, and are addressed in

section ‎4.2, Data Source and Rule Types.


December 2011 20

Data For Analysis

A data source offers several fields for the business rule to use in filtering or finding deficiencies. Since

many business rules might use the same data source (in fact, we recommend that as a good practice),

it is likely that any one business rule might only want to use a few of the fields offered by the data

source. This screen lets you pick the ones of interest, simplifying your downstream tasks.

In the Programmed data source/rule type, this step is skipped, as explained in the following sections.


December 2011 21

Filter Criteria

Of all the business rule fields picked in the previous step, some will be useful mainly in filtering out

data that is not of interest. You should pick such fields as filters, and define filter conditions against

them. This is a standard SAP screen and is common to many applications.

Note: The appearance of the top part of this screenshot above differs from the previous screenshots in this section. This is because previously created rules were used to illustrate the remainder of this document. The tabs correspond one-to-one with the steps in business rule creation, with matching names.


December 2011 22

Deficiency Criteria

A deficiency is a condition which requires human attention. This section of the business rule lets you

define such conditions. There are two main ways to do this: you can treat everything pulled back by

the data source as requiring human review (analysis type Review Required, described below in

section ‎4.2.4, ABAP Report Data Sources and Rules), pick a specific field and define a logical

condition against it (for example, document amount exceeding a set limit). A variation on the latter

would be to define a calculated field deficiency, which represents an arithmetic/logical operation on

any of the available fields. Calculated fields are explained fully in the next section.

For all such deficiency criteria, you can choose a value check or blank check. Blank check restricts

you to say whether a field should be populated or blank. Value check assumes the field has a value,

and allows you to define a wide range of conditions using the usual logical operators such as equal to,

less than, between, and so on.

You can define three conditions, corresponding to three levels of deficiency: low, medium and high.

The Deficiency Description column allows you to configure what to call each deficiency level.

Note that the previous screenshot shows two deficiency criteria. It is possible to have multiple

deficiency criteria, each possibly with their own calculations. The rule interprets these criteria as a

logical inclusive OR: each row of data returned by the data source (and, of course, matching filter

criteria) is evaluated separately by each deficiency criterion. A row of data is judged deficient if any of

the criteria classify it as such.

We recommend that a separate deficiency be defined when there are multiple criteria, each of which

has its own conditions. PC 10.0 reports data rows which match a deficiency condition grouped

together by that deficiency condition, making it easier for control owners to understand the problems

and act upon them. In the previous example, it would have been technically possible to compound the

logical conditions into a single deficiency, but harder for the control owner to subsequently interpret

and act upon.

Conditions and Calculations

Use this tab to define the calculations necessary to compute the value of a ―calculated field‖

deficiency.


December 2011 23

PC 10.0 uses BRF+, the standard Netweaver rule engine, to allow users to define calculations. You

can configure very powerful processing using this rule engine, and the goal was to make it easy to

configure relatively simple rules (calculate an average of two fields, say, or compare two dates), and

yet provide a path to configure more complex rules if needed.

The ―Calculations‖ tab (or corresponding wizard step when creating the rule) allows three types of

calculations: a Field Value calculation, a currency conversion, or grouping and aggregation (as shown

in the following screenshot).

Field Value Calculation

PC 10.0 provides a simplified user interface for relatively simple conditions and calculations, and

advises customers to use the full BRF+ workbench to define more complex calculations. For full

documentation on BRF+ and the BRF+ workbench, consult the Netweaver documentation.

The following screenshot shows the simplified GRC user interface to BRF+ functionality.


December 2011 24

One important restriction is that the definition of a calculated field in the deficiency criteria screen

(above) is one-to-one related to the definition of the calculation itself in the conditions and calculations

tab. This means that any significant computation which requires intermediate variables is too complex

to handle here—it would be necessary to define such complex rules in the BRF+ workbench.

One decision method offered by BRF+ is directly incorporated into the PC rule interface: the decision

table. This is called a ―pattern‖ in the PC 10.0 interface, and is available only for the change log check

category of business rule. The point of using a decision table is to enable multi-field deficiency criteria

via logical combinations. This is easiest to understand with an example—please see section ‎6.2,

Appendix B - Change Analysis Example.

The decision table is standard BRF+ functionality, incorporated into our UI. This is explained more

fully in section ‎4.2.2.1, Change Log Data Sources and Rules.


December 2011 25

Currency Conversion

A key feature of the PC 10PC rule engine is the ability to convert currency amounts. This feature uses

core Netweaver support for currency conversions, and leverages the same underlying currency tables

and features as used in ECC, CRM and other SAP applications.

To use this feature, a deficiency criterion must be of type Amount, and the same must be true of one

of the fields available in the rule.

The following screenshot shows a deficiency field of type Amount. Notice that the deficiency is not

directly one of the rule‘s fields (from the Data for Analysis tab/wizard step); rather, it is a ―calculated

field‖, created by choosing the Calculated Field pushbutton.


December 2011 26

For that deficiency, the next tab (or next step in the creation wizard) lets you define the actual

calculations (as shown in the following figure).

Grouping and Aggregation

The screenshots in the section on Currency Conversion also include grouping and aggregation. The

other deficiency in that example, Total Number of Payments to One-time Vendors, is intended to find

the number of payments made to each one-time vendor, and then apply the configured thresholds to

determine if that violates policies.


December 2011 27

The calculations tab (or wizard step) shows how this is setup.

The grouping is on Vendor number, and the aggregation method used is Count—which simply counts

how many times each vendor (the grouped-by field) appears in payments.

Grouping and Aggregation can also be combined in sequence with other calculation methods. The

deficiency criterion of the previous example looks at the total amount paid to each one-time vendor,


December 2011 28

and it does so by first converting the amount into a standard currency (US Dollars in the example),

and then groups by vendor and totals the (converted) amounts.

Notice that the grouping/aggregation calculation is the second in the sequence, with currency

conversion being first—we want to convert to a single currency before adding.

BRF+ Workbench

To leverage the full power of BRF+, first create a stub PC Business Rule, and use the generated rule

ID (not the name—see section ‎4.1.1.1, Creating Data Sources).

Accessing the BRF+ workbench depends on your system setup (NWBC, Portal, and so on.) For this

document, BRF+ was launched from SAPGUI by using transaction code BRF+. First you must know

the technical ID of the rule you created, which you can see in the following screenshot of the PC

Business Rule finder page. The technical object ID of each rule is displayed in the left-most column.

This technical ID serves as the base, or first part, of the BRF+ rule ID in the BRF+ workbench. The

easiest way to find the corresponding BRF+ rule in the BRF+ workbench is to paste this ID, add the

wildcard character ‗*‘ to it, and then search. In the left-hand panel of the BRF+ workbench screenshot,

there are two BRF+ rules with the same base ID as the PC 10 rule. this is because BRF+ creates new

versions of every such rule each time it is changed. In general, you would want to use the last

version, which would be sequentially the highest number in the list of matching rules.

Having found the rule in BRF+ workbench, you can now enhance it to include any BRF+-supported

logic.


December 2011 29

Note: The above information is provided to enable you to define more powerful processing rules in BRF+ workbench, as needed. This requires care in design and implementation. BRF+ is extensible in many ways, and at the extreme end you will essentially be writing code. Whatever custom processing you define, you must take care not to change the input and output parameters—those are used in the integration between PC and BRF+. you must test the custom BRF+ rule to make sure it does what you want, and that the performance impact is acceptable.

Output Format

This section is common to all business rule/data source types, and arranges the output of any

detected deficiencies in the left-to-right column order specified. You can also hide unwanted columns

here.


December 2011 30

Technical Settings

These primarily affect the execution and performance of monitoring. Most data sources (although not

all) will allow users to cap the maximum amount of data they will process, as a performance

management feature. Since performance can be difficult to predict and manage—too much depends

on the size of tables, network issues, etc.—we strongly advise all customers to test the performance of

any monitoring rules before putting them into production.

Note that most monitoring rules can be run in synchronous or asynchronous mode.

Ad Hoc Query

As you define some types of data sources and business rules, you can run them to test if they are

working correctly. This is useful for configurable business rules and data sources, which are designed

and implemented directly from the PC 10.0 user interface. For programmed, SAP query, ABAP report,

and some other data source and business rule types, ad hoc query is not available, since the key

design and testing activities for those types occur elsewhere, and PC 10.0 mainly reflects the

implementation done elsewhere.

For configurable rules, however, ad hoc queries are very useful indeed. The following screenshots

show two modes of ad hoc query operation: one that collects the data as the data source would, and

another that applies the rule logic to filter the data and then apply deficiency logic.


December 2011 31

Attachments and Links

This part is the same as in Data Sources, described previously.

4.1.1.3 Assigning Rules to Controls

Monitoring rules need to be assigned to local controls. the mechanics of the process are briefly

described here.

The Business Rule Assignment link brings up the following page.

The search widget at the top of this page lets you search for local controls—that is, controls assigned

to a particular organization node. The next step is to select it in the middle part of the screen, by

clicking on its row. You then modify the business rules assigned to it by choosing the Modify

pushbutton, and then choosing the Add pushbutton in the bottom portion of the screen. A screen

displays that allows you to search through Business Rules in the ―Active‖ state, which you can then

assign to the local control. You can also modify existing assignments and maintain frequencies of

monitoring or compliance checks.

Once this assignment step is complete, you will be able to schedule the monitoring rule in the

Automated Monitoring scheduler.


December 2011 32

4.1.2 Runtime

The previous activities described how to set up monitoring rules and associate them with controls.

Aside from ad hoc queries in (some types of) rules and data sources, the Planner and Automated

Monitoring Scheduler provide the only means of running rules.

4.1.2.1 Scheduling

The monitoring scheduler is also on the Rule Setup page introduced in section ‎4.1.1, Design-time,

above. The relevant section looks approximately like the following figure. The exact list of links which

appear in any section depends on the role you have.

Select the Automated Monitoring link. the following screen displays.

Use this page to schedule all schedule-driven rules (once they are assigned to a local control as

described above). The only exception in PC 10.0 is the Event data source and business rule.

The Scheduler page displays all currently scheduled jobs. You can create a new monitoring job by

choosing the Create Job pushbutton, which walks you through the process. The following screenshot

gives an overview.


December 2011 33

The top of the screen shows that scheduling is a 5-step process, and the wizard guides you through it.

The most important thing to note about the scheduler is that you can run jobs as frequently as hourly,

and as infrequently as annually.

4.1.2.2 Monitoring Jobs

Job schedules can be created as shown in the previous section, and schedules, once created, can be

viewed from there as well. But this page focuses on the job itself, rather than the results of any

execution of the job. Results can be seen from the Job Monitor, shown in the following figure.

The results of any job can include sensitive data, and PC 10.0 restricts visibility by users‘ roles. Thus

the ability to see the job monitor does not equate with the ability to look at the detailed results of a job.


December 2011 34

4.2 Data Source and Rule Types

This section describes each of the nine data source and rule types. The goal is to explain the

differences, the unique characteristics of each, and where each will be most useful. This document

does NOT provide step-by-step guidance on how to set up each—that is contained in detailed user

guides published as part of product documentation.

Section ‎6.3, Appendix C: Data Source and Rule Type Summary, summarizes all the subsections

below for a handy reference to data sources and business rules.

4.2.1 SAP Query Data Sources and Rules

SAP Query is a Netweaver query tool. On ABAP backend systems to be monitored, queries

previously defined in the SAP Query UI (SQ01/SQ02) can be invoked from the GRC system via RFC,

and their results gathered and presented to the rule engine in PC 10.0. The following screenshot

shows the transaction SQ01.

The following two screenshots show the relevant sub-scenario for Data Source definitions.


December 2011 35

Conceptually, this is the simplest of the data source types. The remote data source is a query defined

in a well understood query engine, and the results are returned as a result set not unlike what SQL

engines return. Of course, SAP Query runs SAP‘s own OpenSQL. But whether you want to use

either an SAP-delivered query, or define one of your own, the methods are quite widely known.

In defining a data source against a previously-defined SAP Query, the designer has to point to a

particular backend system which is to be monitored. PC looks up the set of available queries in that

backend system (including wildcard searches), looks up the query details, and makes its results

available to the PC 10.0 rule engine.

To create any Business Rule, the first step is always to select the (active) Data Source on which the

rule will operate. Since this fixes the sub-scenario, you do not have to pick the sub-scenario for any

Business Rules—it is always inherited from the Data Source.

For SAP Query Business rules, you can define two categories of business rule, as follows.


December 2011 36

The Exception category means that any data returned by the data source is always considered an

exception. The Analysis Type field decides whether to treat all such exceptions as deficiencies to be

remedied (―Set Deficiency Indicator‖), or as something a human must review to determine if it requires

a remedy.

The other category, ―Value check‖ (not shown), implies that there are deficiency criteria which explicitly

need to be evaluated, and that you will then be expected to configure in the ―Deficiency Criteria‖ and

―Conditions and Calculations‖ steps of the create rule wizard (or the corresponding tabs in a previously

created rule).

Further details can be found in the associated user guide (see section ‎5, Further Reading)

4.2.2 Configurable Data Sources and Rules

A configurable data source defines a query against tables in the monitored ABAP backend system

(such as ECC/ERP, SRM, and so on). This differs from the SAP Query data source (described

previously), which merely references the query logic previously configured in SQ01 on the monitored

system. In this case, the GRC user can completely define the query in the GRC UI, on the GRC

system, and have it execute remotely on the monitored system. This has two advantages: you don‘t

need to modify a production backend system to define it, and you can test and update it without

affecting two different systems: GRC and the monitored system.

This section also explains the Change Log option, which tells PC to reconstruct past configuration and

master data settings over the timeframe of the control, and validates all such past and present settings

against the user-configured monitoring rule.


December 2011 37

Having picked the ―Configurable‖ sub-scenario, you next pick a connector to the backend system

against which you want to define the query (see section ‎4.1.1.1). Next, look up the main table for your

query by searching against the list of tables in the monitored (remote) system. As the following

screenshot shows, you can monitor against Transparent tables. You can also use wildcards to search

against table names or descriptions.

Although not shown in the screenshot below, you can also monitor Pool and Cluster tables.


December 2011 38

Having picked the main table, you can next pick related tables to bring in additional information. PC

10.0 looks up the main table‘s relationships to other tables from the monitored system‘s data

dictionary.


December 2011 39

Again, you can use wild cards to search for tables. Note that PC 10.0 already filters the list of tables

to include only those which have related information. The ―Reference‖ or ―Dependent‖ tables option

define the direction of the relationships: dependent tables are those which refer to (as foreign keys)

the key fields of your main table (primary keys), while reference tables are the opposite—they hold the

primary keys to which your main table refers as foreign keys.

You can join multiple related tables together in such a compound data source, with the constraint that

the join conditions are restricted to being equality relationships between like-type fields. For the most

part, it is expected you will join primary keys to foreign keys. PC 10.0 looks up known relationships

from the data dictionary and pre-populates the join conditions area as you go, as shown in the

following figure.


December 2011 40

As suggested in the previous screenshot, it is possible to go beyond relationships explicitly stored in

the data dictionary, by adding or removing join conditions. But all join conditions are constrained to be

equality conditions between fields of the same type.

Joining tables is unavoidable for any system which uses relational databases, of course. But if the

tables being joined are large, the execution of such a query consumes many cycles, and rule

designers have to be careful in what queries they craft here. Many issues impact performance, so

SAP strongly advises customers to test their monitoring rules for performance before deploying into

production. The tool itself restricts rule designers to joining no more than 5 tables in a single data

source (query), and warns the designer as they approach this limit.

Section ‎6.1, Appendix A - Configurable Rule Example, explains the business logic behind the previous

example, and also shows the corresponding business rule definition, including the use of currency

conversion and grouping and aggregation.

Technical Note: Joins as illustrated above are only possible between transparent tables. Pool and

Cluster tables can be monitored individually, but cannot be joined—not to each other, nor to

transparent tables. This sometimes shows up in unexpected ways: for instance, table T685Z is a

pricing table (ERP SD module), with a pair of pricing limit fields. Those fields are currency fields,

which requires a look up of the corresponding reference table. That reference table, unfortunately, is

not a transparent table (it is actually a structure), and so PC cannot monitor the price limit fields of

T685Z—at least, not via the configurable data source/rule type.

4.2.2.1 Change Log Data Sources and Rules

A change log rule is a variation on the configurable rule defined previously, and hence is presented as

a subsection of that type in this document. It is intended to be used for monitoring configuration and

master data tables only.

SAP applications have extensive change-tracking mechanisms for database tables, which guarantee

that all changes are captured, even if they are of very short duration. That is, even if one change is

quickly followed by another, both changes are reliably logged. These mechanisms cover changes

made directly in the system, and also changes transported into the system. This guarantee is very

important for GRC customers, whose primary goal is to ensure that business processes comply with

policies.

Change Log data sources and business rules parse the data in the change logs of configuration and

master data tables, and reconstruct any past settings over the timeframe of the control. So a change

log business rule allows you to check the validity of a configuration or master data setting at any time,

with confidence that all changes made to that setting will be found and tested for correctness. Wrong

configurations are caught, no matter how transiently they were in effect.

Section ‎6.2, Appendix B - Change Analysis Example, shows a detailed example of such a change log

monitoring rule.

Technical note: Configuration changes can be transported in, and are covered by the Basis change

tracking mechanism). Master data can only be changed directly in the system. Configuration and

master data tables have different change tracking mechanisms, and PC 10.0 parses the correct

information for appropriate table types. Master data uses the Change Document mechanism, while

configuration tables rely on DBTABLOG based change tracking.

Performance Impact of Change Logging

Customers often comment that standard IT practice is to turn configuration table logging off globally,

and turning this setting on raises concerns about performance impacts. This is addressed in more

detail in Note 1653464 - Enable Change Log Monitoring by Activating Table Logging. SAP guidance

on this matter has always been that as long as only configuration/customizing tables are being logged,

and a reasonable purging/archiving strategy is used for DBTABLOG, there should not be any

discernible impact on performance or memory of enabling logging in production systems. ECC ships

https://service.sap.com/sap/support/notes/1653464


December 2011 41

with low-volume tables enabled for logging, so enabling such logging is unlikely to degrade

performance.

Finally, recent customer experience suggests that purging (and, optionally, archiving) the DBTABLOG

table about every 13 months provides customers the benefits of PC 10.0 change analysis, without a

sizeable impact on database size.

Change Tracking: Logs versus Polling

In SAP Basis, every time a table is changed, the change is logged in a table as long as (a) logging is

enabled, and (b) the table being changed is marked for logging. This eases the challenge of change

monitoring: instead of ever-more-frequent monitoring to make sure even transient changes are not lost

(―lost‖ in the sense of not showing up on review), customers can feel secure that they will catch every

single change.

Some other products use polling, in which the table in question is queried frequently, and the state is

compared to prior states to detect changes. But it is only half a solution—if polling shows no change,

that‘s fine. If a change does show up, though, you can‘t be sure that more changes did not happen

between two polling snapshots. With polling, you always run the risk that a fraudulent change was

made, then obscured by another change. Basis change logging will miss no change, no matter how

transient.

For customers who are unable to use Basis change logging, PC 10.0 does support a polling-based

fallback option. But we believe that (a) logging changes via Basis functionality does not impose any

significant burden on the monitored system, and (b) frequent polling may adversely impact the system

being monitored.

If Basis monitoring is not enabled, customers must schedule the polling program as a background job

in the monitored system (delivered with the GRC plug-in), choosing a polling frequency which best

balances the need to catch changes against the impact on performance. If Basis change logging is

enabled after polling has been scheduled, the GRC plug-in recognizes the redundancy and stops

polling. The history reconstruction module of PC 10 then looks at polled snapshots as well as the

change log, so as not to miss any known changes regardless of the tracking mechanism.

Definition of Change Log Rules

Change log based rules (whether using Basis logging or GRC polling) can be based on either

―configurable‖ data sources, or ―programmed‖ ones. (Again, we emphasize that the word

―configurable‖ refers to the fact that users can configure the logic of the query, rather than have to

program it in ABAP or SQL.) Such change-log-based rules can be used to monitor either

configurations (via the DBTABLOG functionality mentioned above), or master data (by a different

document master system of tracking changes). The design of this feature shields the rule designer

from the complexities of log analysis; they simply find the right tables and fields to monitor, and the

GRC code tracks down the additional information needed to reconstruct past settings from change

logs.

Once the data source is ready and active, you can define a monitoring Business Rule on it. It is in the

process of defining the rule that the user must indicate that the rule in question is a change log type

rule. Thus, change log rules can only be defined:

For configuration and master data tables

Based on either ―configurable‖ or ―programmed‖ type data sources


December 2011 42

The rest of the rule definition is similar to configurable rules, with one exception. For change log rules

based on configurable data source types (but not for programmed data source types), PC 10.0

provides an analysis type of ―pattern‖, which allows users define a multi-field deficiency criterion using

a decision table.


December 2011 43

The decision table is BRF+ functionality, and the following figure incorporates BRF+ widgets to define

the contents of the decision table. Refer to BRF+ documentation for details, but note that multiple

rows in the decision table are treated as clauses to a logical ―OR‖ expression. Columns within such a

row are treated as clauses to a logical ―AND‖ expression. For example, use the letter C to denote the

contents of a single cell in such a decision table, with row and column numbers in the subscript. Thus,

C3,4 would refer to the third row, fourth column. Then the decision table is evaluated as:

(C1,1 AND C1,2 AND C1,3…) OR (C2,1 AND C2,2 AND C2,3) OR …


December 2011 44

Table Handlers

When using change analysis, one additional configuration requires additional explanation. When

interpreting the change log, the GRC backend plug-in needs a handler to interpret the change log

entries. Sometimes more than one table handler is registered for the table in question, and it can be

difficult to determine which handler to use. Sometimes you may have to pick between two or three

handlers to determine which one is appropriate. The correct handler for your situation will be the one

which makes your deficiency fields available for use in change analysis rule.


December 2011 45

Further details can be found in the associated user guide (see section ‎5, Further Reading).

4.2.3 BW Query Data Sources and Rules

SAP Netweaver Business Warehouse (BW) has a query tool which you can use to define queries

against BW info objects and info cubes. If you want to monitor your business processes by looking at

data extracted to BW, you can create BW queries. Definition of BW Query data sources is very similar

to SAP Query ones described in section ‎4.2.1, SAP Query Data Sources and Rules.

However, there is one critical difference in the way the BW queries must be set up. Unlike

transactional systems, the whole purpose of data warehouses is to enable users to analyze data

intensively. BW queries can be designed to allow follow-on analytics even after the query returns

results. Automated monitoring cannot use those features, since the recipient of the results in this case

is the GRC product, not directly the user. So the BW query used for PC 10.0 automated monitoring


December 2011 46

must be of a more restricted type. For use with PC 10.0 automated monitoring, BW Queries should be

constructed with the following constraints:

1. BW characteristics of interest must be arranged in the row area in Query Designer. The

characteristics are a list of output fields which will appear as ―data for analysis― in a GRC data

source.

2. BW key figures must be arranged in the columns area in Query Designer.

3. Free characteristics in Query Designer may not be used.

4. Characteristic restrictions should not be used for query filters.

5. Only Single Value and Selection Option variable presentations of query filters should be used.

6. All query filters and selection fields must be optional—GRC monitoring rules do not require

filters to be specified, so the underlying BW query must also make this optional.

7. In Query Designer‘s row area, the result row setting must be Always Suppress.

8. Versions 7.X and 3.X of BW query designers are not supported.


4.2.4 ABAP Report Data Sources and Rules

ABAP reports are widely used to monitor transactions, configurations, master data, and so on, in SAP

applications. PC 10.0 can use such reports for automated monitoring. PC can run such reports on a

user-defined schedule, bind values (for example, company codes, dates) to report parameters, and

present the results to PC users for review. Use of report variants is supported, but not always needed.

Results are viewable as an image, PDF document, or spreadsheet.

ABAP reports have a range of behaviors, some of which are not suitable for PC 10.0 automated

monitoring. PC 10.0 must invoke ABAP reports remotely, so first of all the report must be able to run

as a background job on the monitored system, and its results must be able to be stored as a spool job.

Secondly, it must not use additional popup screens for user input, after its declared parameters are

bound to values. It is difficult to know which reports qualify under these circumstances, so the PC

team has written a program to help qualify ABAP reports for use in automated monitoring. The how-to

guide for this data source and business rule type describes the qualification program and other details

(see section ‎5, Further Reading).

4.2.5 Netweaver Process Integrator-based Data Sources and

Rules

SAP GRC can monitor non-SAP applications, and the Process Integrator-based data source is one of

several suitable for this purpose.

Netweaver Process Integrator (PI, formerly known as Netweaver XI) is SAP‘s preferred platform for

integrations. PI supports common protocols such as JDBC and ODBC, and customers who have

non-SAP applications in their system landscape can use PI to either query directly against the

underlying databases of these applications via JDBC/ODBC, or even integrate at the code level, if

appropriate.

The Monitoring API

PC 10.0 provides a four-method Monitoring API to integrate with PI, grouped into two design-time

functions and two run-time functions. Design-time functions are useful to rule designers, and run-time

functions are used to execute the rules so defined. This section describes the API at a conceptual

level, although the technical names of implementing functions and proxies may well be different. The

descriptions in this section are also relevant to other data source types described below, as noted

there.

The four methods can conceptually be called GetQueryList, GetSignature, ExecuteQuery and

TestQuery. The first two are the design-time methods.


December 2011 47

GRC calls the remote system via GetQueryList to get a list of queries or other data-retrieval functions

on offer from the remote system for use in monitoring. This list is then displayed to the rule designer

to create GRC data source objects.

Once the rule designer has picked a particular query from the list, GRC calls the GetSignature method

to retrieve meta-data about the input parameters the query expects, and the results it returns. These

are reflected in the data source attributes, and become available for designers to use in defining filter

criteria, deficiencies and output formats.

The TestQuery method is used for ad hoc queries from data sources and business rules, although

note that this method is not always available due to technical reasons.

The ExecuteQuery method is called to actually run the query once scheduled, or executed via the

Planner. The key difference between TestQuery and ExecuteQuery is that TestQuery calls are not

logged or audit-trailed, while all scheduled executions are.

PI Implementation of the Monitoring API

PC 10.0 includes a PI implementation of the Monitoring API described previously. So customers or

partners who want to create PI-based queries (for instance, to query directly from a non-SAP

database via ODBC/JDBC) can define such queries in PI, and use the GRC-supplied implementation

of the Monitoring API to integrate those into PC monitoring rules.

PI rules may use asynchronous communication between PC 10 and PI, which requires a callback

endpoint in PC.

The associated user guide describes how to configure PI-based rules in more detail (see section ‎5,

Further Reading).

4.2.6 Web Services-based Data Sources and Rules

These are also called ―External Partner‖ data sources, since they have historically been used by SAP

PC partners to enable easier monitoring of different backend applications. In the IMG configuration

screens, you will also notice the term ―GL-MQT‖, which again reflects this history. PC 10.0

generalizes to web services what used to be proprietary interfaces for such partners to implement.

This API is the same as described above in section ‎4.2.5, Netweaver Process Integrator-based Data

Sources and Rules, with the exception that the Test method is not available for partners to implement.

The method implementations are, of course, the responsibility of the partner product (or custom

middleware). In web services, it is usual for the implementer of a service to publish a WSDL definition

of the service, to which the caller then adapts. In our case, however, PC 10.0 has a fixed outbound

proxy, and we have published the WSDL which the proxy expects the partner to implement. These

are described in SAP Note 1549031. Further details on this data source/rule type can be found in the

associated user guide (see section ‎5, Further Reading).

4.2.7 SoD Data Sources and Rules

The GRC Access Control 10.0 (AC) solution includes extensive rules covering segregation of duty

conflicts, critical actions & permissions and so on. These rules conform to the rules interface of the

PC automated monitoring framework. Such rules can be run in the context of PC local controls, as

described in section ‎4.1.1.3. One key constraint is that the two GRC products—Process Control and

Access Control—should both be installed in the same system.

AC provides one rule to use with PC 10.0 for monitoring, which takes a fixed set of parameters and

returns a single set of results. This can only be used as-is within PC 10.0.




December 2011 48

4.2.8 Event-driven Data Sources and Rules

Event-driven data sources and rules are different from all the others in one respect: they are not

initiated by PC on a schedule (or by the PC user). Instead, some external program or agent must

decide when an event has occurred and needs to be communicated to PC for monitoring and

evaluation.

PC provides a web services interface which a partner product can call to raise such events to PC. PC

must first be configured to expect such an event in the so-called Event Registry (part of IMG

customizing), and the configuration includes a description of the event payload, or the contained

attributes and their data types. When actual events are received, PC first validates the payload (XML)

against the registered event meta-data, and only those events with payloads which pass muster are

processed further. The event interface is a synchronous method, and only success or failure is

communicated back.

Details about how to configure these data sources, how to obtain the WSDL for the web service, etc.

can be found in the associated user guide (see section ‎5, Further Reading).

4.2.9 ABAP Programmed Data Sources and Rules

Some monitoring rules might require complex processing beyond the capabilities of the query and rule

engines described in this document. Such situations can be addressed by writing custom ABAP

programs to deploy on the backend system being monitored. Such programs must conform to the

overall design framework prescribed for automated monitoring, and must be registered in the GRC

plug-in for monitored systems. These are very similar to ABAP reports, but since they adhere to a

pattern prescribed by the GRC plug-in, they can be better leveraged by PC for monitoring. To

emphasize this distinction, we call them ABAP Reports for GRC.

Once again, the conceptual framework is the Monitoring API as described above in section ‎4.2.5,

Netweaver Process Integrator-based Data Sources and Rules. Any ABAP report for GRC must

implement (extend) a defined base class, and the programmer must create an entry in a specified

table with the program (class) name, parameters, etc. The necessary classes and tables are part of

the GRC plug-in for Netweaver ABAP systems, which must be installed in the system to be monitored.

PC 10.0 can then look up the relevant ABAP reports for GRC and their meta-data, and provide PC

rule designers with the information they need.

Details about how to write such ABAP reports for GRC, how to register them with the GRC plug-in

(RTA), etc. are found in the associated user guide (see section ‎5, Further Reading).

5. Further Reading

PC 10.0 documentation on SAP Service Marketplace includes an installation guide, operations guide,

migration guide for customers of previous releases, and detailed how-to guides on each of the data

source/rule types. The user guides for different data source types are being written concurrently with

this document, and the links below represent what is available as of this writing (November 2011).

Install Guide - SAP BusinessObjects Process Control 10.0

Master Guide - SAP BusinessObjects Process Control 10.0

Operations Guide - SAP BusinessObjects Process Control 10.0

PC10 - Configurable Subscenarios for Automated Monitoring

PC10 - SoD Monitoring Subscenario for Continuous Monitoring

PC10 External Partner Integration for Continuous Monitoring

https://service.sap.com/~sapidb/011000358700001377392010E







December 2011 49

Security Guide - SAP BusinessObjects Process Control 10.0



December 2011 50

6. Appendices

6.1 Appendix A - Configurable Rule Example

The SAP ERP purchasing process enables customers to define strict rules for how their company

manages purchasing, to prevent fraud, comply with financial regulations, and so on. Specifically for

our example, the creation of vendors is very tightly controlled, with specific criteria which the system

helps purchasing managers to enforce. But for practical reasons, it is sometimes necessary to bypass

such checks, and to enable this, SAP ERP supports the concept of a ―one-time vendor‖. The idea is

that for an exceptional situation, there is a way to bypass the rules and push through a purchase

quickly.

This sets up a conflict: on the one hand, customers need the flexibility that one-time vendor use

permits; on the other hand, like any other freedom, this flexibility can be abused. Some customers

have policies which limit the use of one-time vendors to a set amount limit, or limit the number of times

a one-time vendor can be used. These requirements are ―outside the system‖; that is, they are not

configured into SAP, nor are they enforced preventively. This makes sense, since if preventive rules

could be defined, then one-time vendors would not be needed at all.

PC automated monitoring provides a solution for this. Customers can set up a monitoring rule which

expresses usage limitation guidelines such as frequency of use or amount limits. The following

example illustrates this.

In section ‎4.2.2, we showed how to build a configurable data source. Picking up the narrative from

there, the following screenshot shows a PC Data Source which helps query the relevant information.

This is a configurable data source, meaning that the query logic driving it is defined in the PC system

itself, although it will run against a remote backend system. It is looks up the main vendor table


December 2011 51

(LFA1), and joins it to the financial table BSAK to look up related payments for each vendor. The join

conditions are automatically populated from the data dictionary as the related table is selected.

You then pick the fields from the two tables you wish to examine, in the lower half of the picture above.

In this example, the interesting fields are company code, clearing date, vendor number and amount

(not all of these fields are actually visible in the figure). As noted previously, the data source can (and

probably should) include more fields than the ones you need for one rule—the idea is to share data

sources across multiple rules.

The next step is to define the business rule. The following screenshots show a previously configured

business rule, so this will vary very slightly from your experience in building a new business rule, as

mentioned earlier.

The previous picture shows how you pick the fields of interest for the business rule you are defining,

from the presumably larger set of fields the data source has to offer.

The next step is to define the filter criteria, as shown in the following screenshot.

The company code parameter should be bound to the corresponding OLSP parameter, so that control

owners for particular organizational units automatically see data only for their own unit. In the

previous picture, we show how to only look at payments made to one-time vendors.

The following picture shows the next tab (and also the next step in defining the rule), where you define

the deficiency criteria.


December 2011 52

For the rule shown here, two ―calculated fields‖ were created (see section ‎4.1.1.2, Creating Business

Rules): the first one will hold the total amount paid to each vendor in the test timeframe, and the other

will count the number of payments made to each vendor. The lower half of the figure above shows the

conditions for determining whether there is a low, medium or high deficiency based on the total

amount paid to the vendor. The following screenshot shows the definition of the other deficiency

criterion, based on total number of payments made to a vendor.

Note that the name of the deficiency is descriptive, to help the control owner understand what the data

is showing. The deficiency description in the bottom half is similarly descriptive.

The next step is to actually define the calculations needed for these ―calculated‖ deficiency fields.

Note again that we chose in this example to calculate the deficiency from the fields on offer from the

data source. If you happen to have a data source whose existing fields already offer the deficiency

criteria you need, your task is correspondingly simpler, since you don‘t need a calculated field: you

simply use one of the existing fields to define your low, medium and high deficiency conditions.

As you can see in the previous picture, conditions and calculations are defined separately for each


December 2011 53

calculated deficiency field. For the amount-based deficiency criterion, it makes sense to first define a

currency conversion step, since we want to be able to set the limits in a specific currency while

evaluating payments made in any currency. The next screen shows the definition of the conversion.

In the this figure, we are telling PC to take the amount in one of the data source fields, ―Amount in

document currency‖, and convert it to US dollars as of the fixed date of 17 October 2011. Alternately,

you could use the ―Clearing date‖ field in each such payment by picking the second option (―By date

field‖ above), or relate it to the date of execution of the rule (the date of rule execution, the first of the

month in which the rule is executed, and so on).

The next step is to add up all payments to a single vendor, having first converted to a common

currency. The following figure shows the definition screen for this conversion.

As you can see, we group by the vendor number, and sum up the individual payments to the vendor.

Note that we have checked the box ―Show Detail‖.

skip next to the ―Ad hoc query‖ tab, to show the results. Running the rule in ―data collection‖ mode, we

see the raw data the data source will present to the rule, below.


December 2011 54

The key to note is that there are several one-time vendors used, with more than one payment to many

such vendors. The amount paid to these vendors is also in two different currencies. It so happens

that a particular vendor is always paid in the same currency in the above picture, but that doesn‘t

matter—PC converts each payment, so would be able to handle the same vendor paid in different

currencies as well.

Executing the ad hoc query again in ―apply rule‖ mode, we see the following results.

In This screenshot, The first column, Sequence Number, shows how many distinct rows of data

remain after grouping. In this figure there are four distinct vendors left, so only four rows—those are

the ones with a number in the first column. For those rows (that is, with a number in the first column),

the far right column shows the aggregated value (the sum, in our example) of payments to that

vendor, converted when necessary into US dollars. For those same rows, the second column shows

the level of the deficiency, as per the condition defined three steps earlier.

The remaining rows in the figure (that is, the rows with no number in the first column) are the ―details‖,

showing the individual payments to that vendor which have been grouped and aggregated in the other

rows (that is, the rows with a sequence number in the first column). This corresponds with the ―Show

Detail‖ checkbox mentioned two steps earlier.

The definition and operation of the other deficiency criterion, on the total number of payments to a

one-time vendor, are similar and are not shown.

6.2 Appendix B - Change Analysis Example

Summary

This section describes how Process Control 10.0 reconstructs past configuration and master data

settings of monitored systems, and enables customers to assert their compliance requirements in a

flexible, user-configurable rules engine. The functionality is shown in the context of an actual

customer example, based on the widely-used Sales & Distribution (SD) module of SAP ERP.


December 2011 55

PC 10 releases a major upgrade in automated monitoring of key backend configurations and master

data settings. As before, PC is able to extract and analyze changes to important configurations and

master data (we‘ll just call them ―configurations‖ for convenience). But where PC 3.0 presented each

logged change individually for evaluation (―Is this change of vendor account number a deficiency?‖),

now it can put together all changes which went into effect together, so that customers can make more

meaningful evaluations. This is important, because typical configurations have multiple fields which

can be changed individually; sometimes, the combination of the fields could be the deficiency

candidate.

This becomes clearer with a specific example. The Sales & Distribution module of SAP ERP has a

credit check configuration setting which gives customers lots of flexibility managing credit settings for

their customers. There are many configurations (many tables and fields can be configured), and all of

these interact in many and subtle ways, as is typical in ERP applications. To illustrate the use of PC

10 monitoring, though, we will focus on just one aspect: configuring the credit check types to be

performed for particular categories of customers for specific company codes/sales areas.

The transaction in question is OVA8 in SAP ERP systems:

The user starts by picking the sales category, or creating a new one:

http://www.sap-img.com/sap-sd/credit-management.htm


December 2011 56

For our purposes, we just select the first one and choose View Details (or Ctrl+Shift+F2).

The Change View: View for Maintenance of Automatic Credit Control screen displays.

Without entering into the subtleties of Sales & Distribution, ―static‖ checks count outstanding orders,

open or delivered (but not yet paid) against a customer‘s credit limit; document value limits credit

extension to sales upto the configured ―maximum document value‖ limit. Customers report that typical

valid settings are complex combinations of these fields. For instance, the following screenshots are

both valid settings, per one major customer:


December 2011 57

That is, customers can be subject to a static check, or be limited on the maximum amount of a single

sale, but not both. As routine maintenance causes SAP customers to change these settings to match

evolving business strategies, the goal of GRC is to ensure that any deviations from acceptable or valid

combinations can be flagged, without bogging the customer down in a flood of false positives.

Using ―F1Technical Details‖ against the individual fields tells us that these settings are to be found in

the table T691F, with field names CMPAA, CMPAB, and so on.

It is a relatively straightforward matter to look up the tables in question, and define the criteria for

monitoring. The credit check setting is compliant with policies if the following condition is true,

deficient otherwise:

(Static_check = true AND status_block <> TRUE) OR

(Document_value_check = true AND Max_doc_value NOT NULL)

Of course, any real monitoring rule will most likely be even longer, but the above condition is enough

to show that the PC 10.0 rule engine can deal with the type of complexity involved. More clauses

won‘t make the condition any more complicated—just longer.

The real challenge, of course, a simple check like this would go against the current state of the setting,

in table T691F and related views. It tells us nothing about the past. This is never going to be fully

reliable, since the monitoring rule can only be executed a finite number of times—what if the setting

was changed to an invalid one for a short period, and then changed again to something that‘s valid?

What companies need is a way to capture every setting that was ever in effect over the timeframe

being monitored (for example, last quarter, or last year), and apply the monitoring rule against such

reconstructed settings over the whole timeframe. Thus, if the PC 10.0 control to which the monitoring

rule is assigned has a timeframe of FY2010, the rule should reconstruct all SD Credit check

configurations for FY 2010 from the change logs, and apply the monitoring rule against those

reconstructed settings.

SAP Basis provides extensive capabilities for logging changes to such configurations. In transaction

RZ11, we look up parameter ―rec/client‖ (see also SAP Note 1653464 - Enable Change Log

Monitoring by Activating Table Logging).




December 2011 58

Separately, in transaction SE11 we look up the table in question (T691F):

Looking at the technical settings (highlighted button) shows the per-table flag which tells Basis to log

changes.


December 2011 59

With these two settings enabled, the ERP backend will log changes to the table in question, and PC

10.0 can monitor the logs to reconstruct past settings.

The next few screenshots describe how to set up and test a monitoring rule for change log analysis of

SD Credit checks.

Navigate to the Continuous Monitoring section of the Rule Setup screen in PC 10.0.


December 2011 60

Create next a Data Source pointing to the T691F table, in a suitablyprepared ERP backend system.


December 2011 61

run it against the ―SM EA5/100‖ system mentioned previously.

next define a Business Rule monitoring this data source. Again, the following screenshots illustrate

this.


December 2011 62

Note that analysis type of ―pattern‖ here refers to the fact that customers can define a deficiency

criterion which is a pattern over multiple fields, not just the single-field deficiency criterion.

The system then prompts the user to define a handler for the table, and helpfully prompts with

appropriate responses. Finally, the pattern itself (such as the one shown in the example) is shown in

the following figure.

Some people find it more convenient to define/edit the pattern via a spreadsheet, which is a feature

offered in the screen above:

Moving to the ad hoc query tab, we can run the rule or underlying query (data source) in a few

different ways:


December 2011 63

6.3 Appendix C: Data Source and Rule Type

Summary

The chart below summarizes the different data source types supported, when to use them, and what

options and limitations exist for building rules with them. For reference, the screenshot below shows

the options as they show up in the PC 10 UI.

Rule Category

A rule runs in one of three possible modes, which PC calls its category.


December 2011 64

Exception category means that all rows of data the rule finds are treated as a problem, without

further analysis.

Value check means that you intend to define logic in the rule which will be applied to each row

(or groups of rows), which will determine if each row (or group) is a problem.

Change log check is used to tell PC that it should reconstruct all known past settings from

Basis change logs, document history or PC-polled snapshots of the configuration or master

data table in question. Change log check is available only for Configurable data sources.

Analysis Type

Changes, Number of changes, and Pattern analysis types are available only for change log

check category. Changes is used when any change is a problem, regardless of the values

involved before and after. Number of changes counts up the number of times a particular

setting changed over the test time period, and allows the user to set numeric thresholds for

deficiencies. Pattern allows you to define deficiency conditions spanning multiple attributes,

and involving compound logical and arithmetic expressions.

Review required analaysis type goes with the Exception category, and tells PC to create and

route an issue with all the data returned by the data source. PC does not categorize the data

as a deficiency, but instead leaves it to the workflow recipient to manually review and decide

whether there is a deficiency.

Set deficiency indicator analysis type goes with the Exception category, and tells PC to create

an issue if any data is returned by the data source. PC then sets the deficiency indicator to

low, medium or high, depending on how you configure the rule.

Monitor value analysis type goes with the Value check category, and tells PC to expect the

user to define a deficiency condition which PC will then use to decide whether any returned

data constitutes a problem.

Num Data Source

Type

Usage Category Analysis Type Comments

1 ABAP Report Use to leverage

suitable ABAP

Reports

already

available

Exception Review

required

The ABAP Report in question

must be checked and

registered in the GRC plug-in

first

Results cannot be processed

by PC, they are only

presented to the user for

review

2 SoD

Integration

Use to invoke

Access Control

risk analysis in

the context of

PC controls

Exception Set Deficiency

Indicator—

results are set

to one of low,

medium or

high.

Access Control and Process

Control must be installed in

the same system

Review

Required—only

manual review

allowed


December 2011 65

3 BW Query Use to invoke

queries against

SAP Business

Warehouse

Exception Set Def

Indicator

BW Query must be defined in

BW system first, and must

have certain limitations

(documented in section BW

Query Data Sources and

Rules‎4.2.3 above)

Review

required

Value

Check

Monitor value

4 Configurable Change

log

Changes Change log rules can really

only reconstruct past settings

for one table, although

additional tables can be used

for filtering. Transparent,

pool or cluster tables can be

queried, although pool/cluster

tables cannot be part of any

join.

Num of

changes

Review

required

Monitor

Pattern

Value

check

Monitor value

5 Event Value

check

Monitor value

6 External

Partner

Exception Set def

indicator

The BRF+ based rule engine

is not available for rules

based on this data source

type. So only simple

deficiency conditions

(involving single attributes,

with simple atomic

conditions) can be defined.

Review

required

Value

check

Monitor value

7 Process

Integration

Exception Set def

indicator

Review

required

Value

check

Monitor value

8 Programmed Exception Set def

indicator

Review

required

Value

check

Monitor value

9 SAP Query Exception Set Def

Indicator

The SAP Query must

previously be defined in the


December 2011 66

Review

required

backend‘s SQ01/02

transaction(s).

Value

check

Monitor value

7. Acknowledgements

The author gratefully acknowledges considerable help from many colleagues. Martin Chen designed

the overall architecture of the automated monitoring features of PC 10.0. Alex Hsu defined much of

the information architecture, and succeeded Martin in architectural oversight. Jackie Wang

implemented a large part of the functionality, and also provided detailed reviews to the author very

promptly. Daniel Chang and Haiyang Yu also provided very useful feedback on the accuracy of the

information presented here, and the draft user guides written by many members of the development

team were sources of very useful information in preparing this document. Michaela Zwinakis and Paul

Davis read the entire document and gave detailed suggestions which have been incorporated into the

final version of this document. Aman Joshi of Price Waterhouse Coopers gave detailed suggestions,

most of which are also incorporated. Jan Gardiner reviewed the draft thoroughly, and gave very

insightful comments; most of her suggestions were incorporated into the published version. Holly

Marrs and Daniel Fowler of Protiviti also gave very helpful feedback.

8. Copyright

© 2011 SAP AG. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice.

Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors.

Microsoft, Windows, Excel, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation.

IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, System z9,

z10, z9, iSeries, pSeries, xSeries, zSeries, eServer, z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server, PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER, OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX, Intelligent Miner, WebSphere, Netfinity, Tivoli and Informix are trademarks or registered trademarks of IBM Corporation.

Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.

Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks or registered trademarks of Adobe Systems Incorporated in the United States and/or other countries.

Oracle is a registered trademark of Oracle Corporation.

UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.

Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are trademarks or registered trademarks of Citrix Systems, Inc.

HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C®, World Wide Web Consortium, Massachusetts Institute of Technology.

Java is a registered trademark of Sun Microsystems, Inc.

JavaScript is a registered trademark of Sun Microsystems, Inc., used under license for technology invented and implemented by Netscape.

SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer, StreamWork, and

other SAP products and services mentioned herein as well as their respective logos are trademarks or

registered trademarks of SAP AG in Germany and other countries.

Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web

Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their

respective logos are trademarks or registered trademarks of Business Objects Software Ltd. Business Objects is

an SAP company.

Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and other Sybase products and services

mentioned herein as well as their respective logos are trademarks or registered trademarks of Sybase, Inc.

Sybase is an SAP company.

All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary.

These materials are subject to change without notice. These materials are provided by SAP AG and its affiliated companies ("SAP Group") for informational purposes only, without representation or warranty of any kind, and SAP Group shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP Group products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty.

SAP Business Objects GRC 10.0 Automated Monitoring...

Documents

Transcript of SAP Business Objects GRC 10.0 Automated Monitoring...