Page tree
Skip to end of metadata
Go to start of metadata


This section explains the alerting capabilities of the SAP BusinessObjects BI platform support tool.  It describes how to understand what alerts are triggered and why, how alerts are configured, and how to customize or create new alerts.


The Alert Summary is the front page of every Landscape Analysis Report.  It contains meta-data about the report instance as well as a summary of the results of any System Alerts, User Alerts, and Complex Alerts that were triggered for this report instance.


Landscape Analysis Report Summary

The Landscape Analysis Report Summary contains all the metadata about this report instance.  Refer to the below table for a description of the various properties.

SAP BI SystemCMS landscape against which the Landscape Analysis was executed
Data DateDate when the data was extracted from the BI landscape
Report Execution DurationThe amount of time it took to extract and analyze the landscape data
BI Support Tool verThe version used to extract the landscape data
PublishedSpecifies whether or not this report instance has been published to SAP Reverse 911 system
OwnerThe user account used to logon to the CMS system for this report


Alert Summary Accordions

Each Analysis type in this Landscape Analysis Report appears as an accordion on the Alert Summary.  Any alerts that were triggered in this report instance will appear under their specific accordion.  This makes it much easier to scan for any critical issues without having to inspect the entire report page by page to know if an alert was triggered.  For each accordion, the number in parenthesis next to the analysis type provides the total number of alerts thrown for that section.

User Alerts vs System/Complex Alerts

In the BIPST 2.1, there are three types of alerts:  

  • System Alerts are alerts whose thresholds and formula logic are defined by the BIPST development team in order to perform more complex checks on the system (for example, a check to make sure all BI nodes in the system are at the same SP / Patch level).  System alerts typically leverage data from many different analysis reports at once to perform a more complex logic evaluation.  Therefore, to insure all system alerts are checked, make sure you run all available analysis types in your Landscape Analysis Report.  System Alerts can be disabled in the Alerting interface.
  • User Alerts are alerts whose threshold can be customized by the user via the Alerting interface.  User alerts allow only one property/metric (e.g # of connections), one operator (e.g. greater than, less than, contains, does not contain, etc), and one threshold value.

  • Complex Alerts are alerts that are customizable and combine two or more alert formulas.  If all of the defined formulas evaluate to true then the complex alert is triggered.  This type of alert is useful for doing more complex evaluations.  For example, you could alert if there is a server that is a jobserver and it is running more than 10 jobs concurrently.


What System Alerts are included?

Refer to the 2 tables below for a list of system defined alerts that are evaluated when executing the Landscape Analysis Report

Alert Description / PropertyThreshold / Formula
AdaptiveJobServer Sizing CheckThis alert will trigger if more than one AdaptiveJobServer is found to exist on a single node.
AdaptiveProcessingServer Split Check

This alert will trigger if any AdaptiveProcessingServer has all 14 default services underneath a single APS server indicating it has not been split to best practices.

Analysis for Office AD SSO problem with Tomcat 8 CheckThis alert will trigger if Analysis Office workbooks are detected on the system, Tomcat 8 is detected and -Dorg.apache.catalina.core.ApplicationContext.GET_RESOURCE_REQUIRE_SLASH=true is not observed in the java options of the Tomcat
APS Initial Heap Size (-Xms) CheckThis alert will trigger if the initial heap size (-Xms) value is set less than 512m on the APS startup command line.  Per the BI Sizing Guide, it is recommended to set the -Xms value on any given APS no less than 512m.
Available Drive Space CheckChecks the space available for each drive on each host and alerts if any drive is < 10GB. Requires Host Agent to be installed on the OS and the Hardware Analysis Report was selected.
BI Add-On Version Consistency CheckThis alert checks for version consistency between nodes for the BI Add-Ons: Explorer, Design Studio and Lumira
BOE Web App Build CheckThis alert will trigger if the BOE web application build number (found in tomcat/webapps/BOE/WEB-INF/internal/version.txt) does not match CMS build number.
CMS Auditing DB Update CheckThis alert will check if auditing is enabled, and if so, verifies that the CMS metric 'Auditing Database Last Updated on' has been updated within the past 5 days. If auditing is enabled and the date is greater than 5 days, this alert will be triggered.
CMS Horizontal Sizing CheckThis alert will trigger if there are 3 or more nodes and all nodes contain a CMS server.
CMS User Sizing CheckThis alert will be triggered if too many cms's exist based on the 'peak user sessions' metric. This only evaluates if 3 or more cms's exist and if the following formula is true: ((CMScount * 500) - peak users) < 500
Corba SSL Memory Leak CheckThis alert checks if the environment is susceptable to a BI4.2 memory leak found in the SIA process when corba SSL is enabled.
Inconsistent Server StateThis alert looks for servers in an inconsistent running state. The alert triggers when a server is marked as 'running' but does not have a start time metric.
License ValidationThe 'License Validation' System Alert reads the license keys added into the CMC and checks the expiration date.  If the expiration date is set to expire within 30, 60 or 90 days you will be alerted accordingly.
Lumira Server Sizing CheckThis alert will trigger if any enabled BI server is found on the same host as the Lumira APS.  Sizing recommendation is such that it should be placed on a dedicated server.
Patch ComparisonThis will compare the patches on each host. If all nodes do not have the same patch level, an alert is thrown.  This alert requires that the Patch History Analysis was selected prior to running the Landscape Analysis Report.
Patch History Root User CheckThis alert checks each patch installed on each host by reading the install.manifest file. Alert is triggered if any BI install step was run by the 'root' user.
Platform Search Corruption CheckThis alert triggers if Platform Search is NOT set to 'scheduled crawling' and a recurring scheduled instance exists for the Platform Search program file.  This combination has been known to corrupt the Lucene index used by Platform Search.
Platform Search Inconsistent Status CheckThis alert is triggered if the Platform Search Application shows a status that does not match the status shown in each APS Search Servers metrics.
Platform Search Master Index Location CheckThis alert triggers when 2 or more Platform Search APS servers exist and a unc path is NOT specified for thePlatform Search Master Index location.
Platform Search Surrogate CheckThis alert will trigger if using BI 4.2 SP04, indexing is set to anything other than Platform Metadata and the patch level is less than 3. See Note 2500438 for more information
Server IP Binding CheckThis alert checks that each BI Server under a given node is bound to the same ip address.
Server with Missing Services CheckThis alert is triggered if any server is found to not contain any services.
Timeouts Check - CR Proc & CR  Cache Idle Connection comparisonThis compares the CR Processing and CR Cache servers Idle Connection Timeout to ensure the values match.
Timeouts Check - WACS & CR Cache comparisonThis compares the RestfulWebService property 'Enterprise Session Token Timeout' and compares it to each CR Cache Server property 'Security Cache Timeout'. The WACS property should be set greater than the CR Cache Server property.
Total Xmx vs Total RAMThis alert will sum up the Java Max Heap (-Xmx) value set on the command line for each java based BI server and compare this to the Total RAM on the same host.  If the total Xmx value exceeds the total RAM value, this alert is triggered.  This alert will only look at the BI servers running on each host. Tomcat, other application servers or java processes running on the same host are not included.
Trace Level (BO_Trace.ini) CheckThis alert checks the trace settings found in the bo_trace.ini. Alert is triggered if 'trace_level' is set to any value other than the default
Webi Memory Analysis CheckThis alert will cycle thru all Webi Processing Servers to ensure the server property 'Enable Memory Analysis' is set to recommended best practices based on the BI version found. This alert will trigger when this property is enabled on BI4.1 SP7 and newer (including BI4.2). In addition, it will trigger if the property is found to be disabled on BI4.1 SP6 and older. For more information, refer to KBA 2276693 and Note 2203828
Webi Processing Server - Servers per Node Sizing CheckThis alert triggers if more than 2 Webi Processing Services exists on a single node. A MEDIUM warning is thrown if 3-4 WIPS are detected. A HIGH warning if > 4 are detected.


Analyzing triggered Alerts

For each triggered alert, there are a number of columns that provide information about that alert.

Target (Metric/Property)The fully qualified path to the alert object in the alert definition hierarchy
Property NameThe property whose value is evaluated against the specified threshold
Property ValueThe value that was found for the property that triggered the alert
Alert DescriptionThe description that provides information about why the alert was triggered, what server (if applicable) is associated with the alert, and any actions or KB Articles and documentation that may help you to resolve the issue
Mouse over the symbol to get more details about the alert definition used to evaluate this alert
AlertSymbol used to indicate an alert was triggered. This symbol will also be found in the specific analysis tab where the property that triggered the alert is located

Alert Details

To get information about the Alert definition for a triggered alert, mouse over the symbol.  This information provides the Alert Name, the Metric description, and the formula used when evaluating the alert.

Customizing Alerts and Creating New Alerts

To customize existing alerts or to create new alerts refer to: 11. Alerting


  • No labels