IBM Performance Management

Advanced Configuration

Use the Advanced Configuration page to control communications settings and advanced features such as event forwarding.

After you click icon System Configuration > Advanced Configuration, the following configuration categories are displayed in the Advanced Configuration page.
UI Integration
For products that integrate with the Performance Management console, you can add or edit the URL for launching the integrated application. The fields are populated with any URLs that were set during the integration configuration procedure.
  • Cognos Reporting URL is used to launch the IBM® Cognos® Viewer for viewing the Performance Management reports through icon Reporting > Work with Reports and Actions > Launch to Reports.
    Note: The Cognos Reporting URL must adhere to the following format:

    https://<reporting-server-hostname>:<port>/tarf/servlet/dispatch/secured

    If you want to view reports using Tivoli® Common Reporting 3.1.1., the Cognos Reporting URL must adhere to the following format:

    https://<reporting-server-hostname>:<port>/tarf/servlet/reporting/secured

    For more information, see Reports.
  • Cognos Reporting Requires XSS Validation is used to reformat the URL for the Cognos report to pass XSS validation. Set this field to True if your configuration enables the Cognos Application Firewall (CAF) and third party XSS checking. Default: False. For more information, see Integrating with Tivoli Common Reporting.
  • Log Analysis URL is used to launch IBM Operations Analytics - Log Analysis for searching through application logs from the Application Performance Dashboard. For more information, see Searching log files.
  • Enable Subnode Events, for agents with subnodes, controls whether subnodes are shown in the Events tab. When subnode events are enabled, the node and subnode for which an event was opened are displayed. Specifically, if you want to display log file monitoring situations in the Events tab, you must ensure that subnode events are enabled. Default: False.
  • Dashboard Refresh Rate controls the frequency of the Application Performance Dashboard automatic refresh. You can adjust the setting to any value from 1 to 60 minutes. The setting affects resource status that is displayed in the navigator and Status Overview tab. It has no affect on the Events tab entries. Default: 1 minute
Event Manager
The Event Manager controls the flow (through forwarding to Simple Mail Transfer Protocol) and storage of received events, and email notification setup.
  • Event Cache Time determines the number of minutes that events are retained in the local cache, as few as 10 minutes or up to 480 minutes (8 hours). If the Event list is long, consider reducing the number of minutes that are kept. Or, to see events over a longer period, increase the number of minutes. Default: 60 minutes.
  • Pure Event Close Time determines how long to leave a pure event open before closing it. Default: 24 hours.
  • Master Reset Event controls whether to send a master reset event to clear previously received sampled events after the monitoring agent is restarted. If you have email notifications configured, you also receive an email (see sample at Master reset event after agent restart). Default: True.
  • EIF Port specifies the port number to use to receive Event Integration Facility events from Operations Analytics - Predictive Insights. If you do not specify a port number for an EIF target in the EIF Event Target(s) field, the value of the EIF Port field is also used as the port number when forwarding events to that EIF target. Default: 9998.
  • EIF Event Target(s) specifies the list of host names or IP addresses and port numbers where all received Event Integration Facility events are forwarded. For example, if you are forwarding events to the Netcool/OMNIbus Probe for Tivoli EIF, enter the fully qualified host name or IP address of the computer where the probe is installed, followed by the receiver port number. Separate the host name and port number with a colon (:), and separate each host name with a comma (,), such as 9.87.65.111:8989,localhost:9090,myhostname.en.ibm.com. If you do not include a port number with the EIF event receiver host name, the value of the EIF Port field is used. IPv6 addresses are not supported for EIF transmission.
  • Target Email Addresses specifies the email addresses that events are forwarded to. Separate each address with a comma (,), such as annette@ibm.com,jim@ibm.com,owen@ibm.com.
  • Email Subject Line is the text to use in the subject line of the forwarded SMTP event email. Default: Performance Management Event.
  • Sending Email Account is the email address to use when sending an SMTP email event.
  • Sending Email Account Password is the password that is associated with the sending email account.
  • SMTP Server Address is the fully qualified host name of the SMTP server that is used for sending events as emails, such as smtp.gmail.com.
  • Use SSL determines whether to use SSL as the SMTP (Simple Mail Transfer Protocol) transport mechanism. Default: False.
  • SMTP Port is the port number to use for sending SMTP email. Default: 25.
  • SSL SMTP Port is the port number to use for sending SMTP email using the SSL (Secure Socket Layer) protocol. Default: 465.
If you configure the SMTP forwarder to use SSL, you must add the SMTP Server’s signing CA certificate to the Performance Management server keystore. Add the CA certificate to the default keystore using the JVM keytool command:
install_dir/java/jre/bin/keytool -importcert\ 
-noprompt \
-alias your_CA cert_alias \
-file path_to_your_CA_cert_file (*.cer) -keystore /install_dir/wlp/usr/servers/min/resources/security/key.jks \
-storepass ccmR0cKs! \
-storetype jks \
-trustcacerts

For information about forwarding your events to the IBM Netcool/OMNIbus event manager, see Integrating with Netcool/OMNIbus.

MongoDB Configuration
Configure the MongoDB database.
  • Host Name is the host name of the system where MongoDB is running. Default: localhost.
  • Port Number is the port number for the Mongo database. Default: 27000.
  • Batch Size is the number of AAR documents that are processed and inserted by the MongoDB ETL into the MongoDB document in a single batch insert. Default: 100.
  • Batch Wait is the time since the last message was processed that the MongoDB ETL waits before an incomplete batch of AAR documents is inserted into the MongoDB document. Default: 10 seconds.
Agent Central Configuration
The settings here are required information that is used by the Performance Management server to configure the Central Configuration Services.
The Central Configuration Services are used to distribute events and thresholds, and historical data settings to the monitoring agents. Each agent of the same type is given the same thresholds and settings.
  • Host Name Override is an optional override for the host name or IP address of Central Configuration Services if the value of the server is not correct. By default, the IP address or host name that you provided when you configured the agent packages is used.
  • Change Check Interval controls the frequency, in minutes, that the monitoring agents use to query configuration details from the server. Default: 5 minutes.
  • Protocol establishes whether HTTP or HTTPS is the protocol to communicate with the monitoring agent. If the APM_SECURE_COMMUNICATION variable was set to Y before server installation, this field is set to HTTPS for a secure connection. For more information, see Setting HTTP or HTTPS communications. Default: HTTP.
  • HTTP Port determines the port to use for HTTP communications. Default: 80.
  • HTTPS Port determines the port to use for secure HTTPS communications. Default: 443.
Data Mart
These settings can be adjusted to improve performance:
  • Batch size is the number of rows to include in a single database transaction. Default: 100 rows.
  • Batch Interval is the interval between sending data to the database to allow it to be inserted. Default: 10 seconds.
Configure an external database for the data mart.
  • JDBC Optional can be used to set extra properties, such as encryption. The format is key=value, one entry per line.
Hybrid Gateway
Configure the IBM Performance Management Hybrid Gateway for viewing monitoring data from your IBM Tivoli Monitoring environment in the Performance Management console.
Important: The following Managed System Group Name, Portal Server Host Name, and Portal Server User Name parameters are mandatory fields that you must populate to configure the Hybrid Gateway. If you are no longer using the Hybrid Gateway later, you cannot set these values to blank values. If you want to indicate that the Hybrid Gateway is no longer required, enter text values in these fields such as "not in use".
  • Managed System Group Name is the Tivoli Enterprise Portal Server managed system group that you created for viewing supported monitoring agents in the Performance Management console. Any monitoring agent types that are not supported by your Performance Management product are not shown in the console regardless of their inclusion in the managed system group.
    For guidance and limitations when creating the managed system group for hybrid enablement, see Creating the managed system group.
  • Portal Server Host name is the Tivoli Enterprise Portal Server host IP address or fully qualified domain name.
  • Portal Server Port is the port number used by the Tivoli Enterprise Portal Server for HTTP communications. The default port is 15200 for HTTP or 15201 for HTTPS. A value of 0 sets the port to the default 15200 for HTTP or 15201 for HTTPS.
  • Portal Server Protocol determines whether to use the HTTP Internet protocol or the secure HTTPS protocol to connect to the Tivoli Enterprise Portal Server. Default: http.
  • Portal Server User Name is the logon user name for starting the Tivoli Enterprise Portal client. This user ID must have access to all monitoring agent types (Allowed Applications is set to All Applications). For more information, see Administer Users in the Tivoli Monitoring Knowledge Center.)
  • Portal Server User Password is the password that is associated with the Tivoli Enterprise Portal logon user name.
  • Pass-Through Proxy Host Name is used if the Tivoli Enterprise Portal Server communicates through a pass-through proxy server. Enter the IP address or fully qualified name of the proxy host system.
  • Pass-Through Proxy Port is used if the Tivoli Enterprise Portal Server communicates through a pass-through proxy server. Enter the port number for communicating with the proxy.
  • Pass-Through Proxy Protocol is used if the Tivoli Enterprise Portal Server communicates through a pass-through proxy server. Enter the protocol used for communications through the proxy. Default: http.
The Tivoli monitoring agents that you are viewing in the Performance Management console are in your IBM Tivoli Monitoring domain. You can view them in the application dashboards, but you cannot create eventing thresholds for these monitoring agents in the Threshold Manager.
Kafka Configuration
The Kafka message broker has predefined configuration settings. Leave these settings at their defaults unless instructed to change them by IBM Support.
  • Producer Batch Size is the number of messages batched at the producer before dispatch to the event handler. Default: 200 messages.
  • Producer Batch Wait is the producer wait time. Default: 5 seconds.
  • Consumer Batch Size is the consumer batch size is the number of records to include in a single database transaction. Default: 100 records.
  • Consumer Batch Wait is the wait time between sending data batches to the database to allow for batch insertions. If the Consumer Batch size is not reached in this interval the data is flushed. Default: 10 seconds.
  • Consumer Thread Count is the consumer thread count. Default: 2 threads.
Tracking Analytics Service
The settings that are used for the Tracking Analytics Service. The settings apply only to the IBM Application Performance Management offering and if you are configuring transactions tracking in your environment.
  • Connection Pool Size is the number of concurrent DB2® connections that the Tracking Analytics Service holds in the connection pool for the Top N query. Increase this value if you are experiencing slow query times due to a large number of concurrent Performance Management console users. Default: 10.
  • Pseudo Nodes enable the visualization of services that are not instrumented. Default: True.
  • Query Timeout in Seconds is the number of seconds before each "Top N" (where N is a number, as in Top 5 or Top 10) query take before timing out. The timeout value can be increased if needed for applications that have a higher workload and have larger expected query times. Default: 120 seconds.
  • DB2 Query Re-optimization Enabled should not normally need to be changed. The parameter affects the DB2 query optimizer. In some environments, turning off the optimizer might improve performance for some sets of transactions. Default: True.
Agent Subscription Facility
The Agent Subscription Facility includes the agent REST (Representative State Transfer) interface and Central Configuration Services HTTP server. The REST interface is used by agents to send monitoring data that is persisted in the DB2 server and threshold events. The Central Configuration Services HTTP server handles requests from agents for their configuration files, such as threshold definitions.
Use these parameters for configuring the communications between the Agent Subscription Facility and the Performance Management server.
  • Missed Poll Limit (Fast Heart Beat) is the maximum number of times that a monitoring agent with a 60-second or lower heartbeat interval fails to connect before it is marked offline. Default: 15 intervals.
  • Missed Poll Limit (Slow Heart Beat) is the maximum number of times that a monitoring agent with a heartbeat interval that is greater than 60 seconds fails to connect before it is marked offline. Default: 3 intervals.
  • Transaction Time Out is the amount of time, in seconds, that the server waits for a response to a request. Default: 120 seconds.
  • Pacing Count is the number of times that the server makes a request to the monitoring agent to call back after a short delay before allowing it to close the connection. Default: 5.
  • Pacing Delay is the amount of time, in milliseconds, that a monitoring agent waits before making a request to the server for more work. Default: 600000 milliseconds (10 minutes).
  • Remove Offline System Delay determines the number of minutes to wait before removing the display of a managed system that is offline. In the Application Performance Dashboard, offline managed systems are indicated by the unknown status indicator. The managed system continues to display, even if you uninstall the agent, until the delay time has passed. For more information, see Viewing and removing offline agents. Default: 5760 minutes (4 days).
Thresholds Enablement
Your monitoring agents each come with a set of predefined thresholds that are enabled and started with the agent. These predefined thresholds are assigned to the default system resource group for the agent. Set the field in the Thresholds Enablement configuration category to Disable All if you do not want to run the predefined thresholds. The Disable All setting removes the assignment of the system group from all predefined thresholds. A threshold with no group assigned is distributed to no monitored systems and remains stopped until it is distributed to a resource group. If you decide later that you want to turn on the predefined thresholds, return to the Advanced Configuration page and set the field to Enable All.
For more information about the predefined thresholds and custom thresholds, see Thresholds background information and Examples of disabled thresholds.