This document enables users to troubleshoot common connectivity issues encountered during PDC configuration for on-premises systems.
Introduction
Pega Diagnostic Center (PDC) on-premises systems running on multiple stacks and networks with different settings and restrictions can cause connectivity issues when pushing data to the Pega Cloud hosted PDC service.
You might have to capture basic log information with the REST/SOAP Connectivity Logger to troubleshoot some of the issues in this support document.
Before enabling Debug mode, ensure that you complete all the steps from Configuring on-premises systems for monitoring with PDC and meet the following prerequisites:
- Each system has a unique name
- All monitored nodes have unique node IDs
- SSL certificates are installed correctly
Enable the following loggers in Debug mode:
- com.pega.pegarules.priv.util.logging.SOAPAppenderPega
- httpclient.wire.content
- com.pega.pegarules.integration.engine.internal.connect.rest.RESTConnector
- httpclient.wire.header
For more information about logger settings, refer to Configuring logger settings.
Common connectivity issues
Firewall rule needs DNS names or IP addresses
Symptom
You are unable to activate outbound traffic to PDC DNS names or specific IP addresses while using a firewall at the network level for communication over the internet to other systems.
Solution
To retrieve static IP addresses, perform the following actions:
- In the command prompt, enter: nslookup <pdc_production_system_name>-service.pegacloud.com
For example:nslookup pdcna1-service.pegacloud.com nslookup pdc-external-service.pegacloud.com
- Record the two static IP addresses provided by the command prompt.
For details about your DNS name, contact Pega Support.
UnknownHostException occurs while connecting to PDC
Symptom
The system displays UnknownHostException
errors in the logs while connecting to PDC after configuring a proxy server at the network level for communication.
Solution
- Set the following parameters in the Java virtual machine (JVM) arguments:
- -Dhttps.proxyHost
- -Dhttps.proxyPort
- If any of the internal endpoints must be connected without a proxy, set the -Dhttp.nonProxyHosts parameter.
For more information about enabling monitoring data, refer to Configuring on-premises systems for monitoring with PDC.
Unable to use application gateway or ESB with PDC
Symptom
You are unable to connect to PDC while using an Enterprise Service Bus (ESB).
Solution
The two following settings are used for PDC connectivity:
- prconfig/pdcconfig/altSOAPURL/default
- prconfig/pdcconfig/altRESTURL/default
To connect to PDC by using ESB, create a custom ESB service that provides mapping.
To connect to PDC through an application gateway, perform the following actions:
- Point altSoapURL and altRestURL Dynamic System Settings (DSS) to the AppGateway/ESB endpoints.
For example:altSoapURL = https://<company_url>/ServiceID altRestURL = https://<company_url>/ServiceID
- Establish a service in the gateway, for example, PDCSOAP. Replace the endpoint with the SOAP endpoint URL and redirect the request to the PDC SOAP endpoint.
- For REST services, set another parameter for the service name after the altRestURL ServiceID instance that is handled in the code.
- Establish a service in the application gateway, for example, PDCREST. Construct the endpoint based on the initial part of PDC REST endpoints and the service name passed.
For example:https://<pdc_host_name>/ServiceID/PegaAES/v1/logAgentInfo
- In the ESB, construct the URL as in the following example:
https://<pdc_host_name>/prweb/PRRestService/<hash>/PegaAES/v1/logAgentInfo
Log4j2-AsyncAppenderEventDispatcher-2-HEALTH-ASYNCAES ERROR
Symptom
The system displays the following log error when connecting to PDC:
Log4j2-AsyncAppenderEventDispatcher-2-HEALTH-ASYNCAES ERROR SOAPAppender _append() failed: java.lang.IllegalArgumentException: The XMLInputFactory does not recognize the property "reuse-instance". java.lang.IllegalArgumentException: The XMLInputFactory does not recognize the property "reuse-instance".
Explanation
A bug occurs in the Axiom JAR files when the monitored node is running on certain IBM Java versions.
Solution
To prevent this issue from occurring, prefix the classpath with other JARs by performing the following actions:
- In the server.xml file, enable the jaxws and jaxb version 2.2 features by adding the following lines:
<feature>jaxws-2.2</feature> <feature>jaxb-2.2</feature>
- Save the server.xml file and restart the server.