You should perform which three steps when troubleshooting a Cisco Unified Contact Center Express engine...

To troubleshoot a Cisco Unified Contact Center Express engine “Java out of memory” crash, you should perform the following steps:

Collect the thread dump for Cisco Unified CCX Engine: A thread dump is a snapshot of the state of all the threads that are part of the process. The thread dump will show what each thread is doing, what resources they are holding, and what other threads they are waiting for. This can help identify any deadlocks, loops, or hung threads that could cause memory leaks or high memory usage. To collect the thread dump for Cisco Unified CCX Engine, use the Cisco Unified Real-Time Monitoring Tool (RTMT) and navigate to Trace and Log Central > Collect Files > Select Cisco Unified CCX Engine > Check Thread Dump > Finish. For more information on how to collect the thread dump, see the Cisco Unified Contact Center Express Serviceability Administration Guide, Release 12.5 (1), Chapter: Unified CCX Serviceability, Section: Trace and Log Central, Subsection: Collecting Thread Dumps.

Collect engine heap performance data via the Cisco Unified Real-Time Monitoring Tool: The engine heap is the area of memory where the Cisco Unified CCX Engine stores the objects it creates and uses. The engine heap performance data will show how much heap memory is allocated, used, and freed by the engine, as well as the number and frequency of garbage collections. This can help identify any memory leaks, memory fragmentation, or insufficient heap size issues that could cause memory exhaustion or degradation. To collect the engine heap performance data, use the RTMT and navigate to System > Performance > Open Performance Monitoring > Select Cisco Unified CCX Engine > Select Heap Usage and GC Counters. For more information on how to collect the engine heap performance data, see the Cisco Unified Contact Center Express Reporting Guide, Release 11.6(2), Chapter: Unified CCX Engine Heap Usage Monitoring.

Talk to the customer about the deployment and usage pattern: The deployment and usage pattern of the customer will provide information about the configuration, load, and behavior of the Cisco Unified CCX system. This can help identify any factors that could contribute to the memory issues, such as the number of applications, scripts, sessions, agents, teams, skills, CSQs, triggers, historical reports, etc. The deployment and usage pattern can also help determine the best practices and recommendations for optimizing the system performance and avoiding memory problems. To talk to the customer about the deployment and usage pattern, use the Unified CCX 12.5 Sizing Tool and the Cisco Unified Contact Center Express Solution Reference Network Design, Release 12.5 (1) as references.

The following steps are not required or not possible to troubleshoot a Cisco Unified Contact Center Express engine “Java out of memory” crash:

Check to see if the customer has installed any third party applications: This step is not required, because the Cisco Unified CCX platform does not support the installation of any third party applications. The platform is a closed system that only allows the installation of Cisco-approved software and patches. Any attempt to install third party applications on the platform will cause the system to fail the integrity check and prevent the system from functioning properly. For more information on the platform compatibility and restrictions, see the Cisco Unified Contact Center Express Software and Hardware Compatibility Guide, Release 12.5(1), Chapter: Platform Considerations, Section: Platform Compatibility.

Collect heap dumps via the Cisco Unified Real-Time Monitoring Tool: This step is not possible, because the RTMT does not provide the option to collect heap dumps for the Cisco Unified CCX Engine. A heap dump is a snapshot of the memory allocation of the process, which shows the objects, classes, and references in the heap. A heap dump can be useful for analyzing the memory usage and finding memory leaks, but it requires a tool that can capture and read the heap dump file. The RTMT can only collect and view trace files, logs, and performance counters for the Cisco Unified CCX Engine. To collect heap dumps for the Cisco Unified CCX Engine, a different tool or method is needed, such as using the jmap command or the [Java VisualVM] tool.

Check the Cisco Unified CCX Serviceability Control Center: This step is not helpful, because the Cisco Unified CCX Serviceability Control Center only shows the status and allows the control of the Cisco Unified CCX services and components. It does not provide any information or tools for troubleshooting memory issues or crashes. The Control Center can be used to start, stop, or restart the Cisco Unified CCX Engine service, but this will not resolve the root cause of the memory problem. To troubleshoot memory issues or crashes, other tools or methods are needed, such as using the RTMT, the CLI, or the web interface. For more information on the Cisco Unified CCX Serviceability Control Center, see the Cisco Unified Contact Center Express Serviceability Administration Guide, Release 12.5 (1), Chapter: Unified CCX Serviceability, Section: Control Center - Network Services. References:

Cisco Unified Contact Center Express Serviceability Administration Guide, Release 12.5 (1)

Cisco Unified Contact Center Express Reporting Guide, Release 11.6(2)

Unified CCX 12.5 Sizing Tool

Cisco Unified Contact Center Express Solution Reference Network Design, Release 12.5 (1)

Cisco Unified Contact Center Express Software and Hardware Compatibility Guide, Release 12.5(1)

jmap

[Java VisualVM]

[Deploying Cisco Unified Contact Center Express (UCCXD) v6.0]8, Module 4: Cisco Unified CCX Maintenance, Lesson 1: Cisco Unified CCX Serviceability and Maintenance, Topic: Troubleshooting Cisco Unified CCX