Skip to Content
Information

This site will undergo maintenance on Friday, 26 September at 2:00 AM CDT / 12:30 PM IST and may experience a short period of instability during that time.

Troubleshooting performance issues after migrating to a containerized deployment

Use the information in this topic to identify and resolve performance issues that occur after migrating from a virtual machine–based deployment to a containerized deployment of BMC Helix Service Management. This topic provides guidance for diagnosing issues related to the database, network, infrastructure, and application performance. It also includes best practices to help you maintain system stability and efficiency after migration. 

Issue symptoms

You may experience one or more of the following performance issues after migration: 

  • Slow system response 
  • Long-running Application Programming Interface (API) requests or Structured Query Language (SQL) queries 
  • Delay in application plug-in performance 
  • Delays during report generation

Issue scope

  • Can affect database, network, or application services.
  • May impact all users or specific groups depending on configuration.
  • Often occurs post-migration or during high load.
  • Can be intermittent or constant depending on resource allocation and configuration.

Infrastructure considerations

Review the following infrastructure requirements to ensure proper configuration and optimal system performance.

  • Ensure that the load balancer is configured correctly for Secure Sockets Layer (SSL) offloading.
  • Confirm that all pods, services, and databases are located in the same data center or region.

Resolutions for common issues

Symptom

Action

Long-running queriesContinuously monitor and terminate any queries exceeding the performance threshold.
Inefficient query performanceAnalyze execution plans, SQL wait statistics, and use of bind variables.
Index fragmentationGenerate the fragmentation report and rebuild indexes.
Latency in plug-insReview CAI logs and compare configurations between environments.
Delayed reportingCapture thread dumps from the Pentaho Carte server for reporting.

Diagnosing and reporting database issues

Perform task 1 to troubleshoot the issue. If you are unable to resolve the issue, perform tasks 2 to 4 to create a BMC Helix Support ticket.

Warning
Important

Before you begin, ensure that you have database administrator access and that the database host has adequate resources such as Central Processing Unit (CPU), memory (RAM), and Input/Output Operations Per Second (IOPS).

Task

Action

Steps

Reference

1

Troubleshoot and self-resolve

Check database host resources and configuration.

Optimize indexes, update stats, and terminate long-running queries.

  1. Ensure that sufficient CPU, memory, and IOPS are allocated to the database server.  
  2. Verify the configuration settings.
    • For Microsoft SQL Server: 
      • Review the Maximum Degree of Parallelism (MAXDOP) and Cost Threshold for Parallelism. 
    • For PostgreSQL: 
      • Review the max_worker_processes and effective_cache_size parameters. 
      • Check vacuum thresholds and last vacuum time. 
    • For Oracle: 
      • Ensure that session trigger settings such as NLS_COMP and NLS_SORT are configured properly. 
      • Confirm that the CURSOR_SHARING parameter is set to FORCE. 
      • Verify that bind variables are being used appropriately. 
  3. Run UPDATE STATISTICS or enable the AUTO_UPDATE_STATISTICS feature in Microsoft SQL Server database. 
  4. Generate index fragmentation reports and perform defragmentation as required. 
  5. Monitor and terminate long-running queries that exceed a defined threshold by using hygiene scripts.
2

Collect diagnostics

Gather logs and query plans.

  1. Review Long Running API (LRA) and Long Running Query (LRQ) reports to identify inefficient queries and missing indexes. 
  2. Check Row-Level Security (RLS) implementations for potential optimization opportunities.
  3. Collect the information to report it to the BMC Helix Support.
For more details on Performance Slowness Due To Long Run Update Query contact BMC Helix Support.
3

Report

Escalate with diagnostics to BMC Helix Support.

  1. Schedule regular statistics collection for all supported databases.
  2. Include diagnostics, LRA/LRQ reports, and system logs in the report.
  3. To raise a support ticket, refer to Submit & Track Support Cases - BMC Documentation
 
4Apply fix from Support
  • Apply the fix provided by BMC Helix Support.
 

Diagnosing and reporting network issues

Perform task 1 to troubleshoot the issue. If you are unable to resolve the issue, perform tasks 2 to 4 to create a BMC Helix Support ticket.

Task

Action

Steps

Reference

1

Troubleshoot and self-resolve

Check connectivity between services.

Adjust Domain Name System (DNS) or firewall settings. 

  1. Measure latency and bandwidth between the platform services and Elasticsearch. 
    From platform pods, use the ping command to check round-trip latency to the database server. Latency should be under 1 millisecond.  
    Important: 
    The ping utility is available within platform containers starting from version 25.x.
     
  2. Use the curl command from inside the pod to perform SOAP or REST-based service calls. 
  3. Validate that there is no packet loss. 
  4. Check for firewall restrictions and proper DNS resolution. 
  5. Ensure correct internal DNS mappings.
 
2

Collect diagnostics

Collect connection and latency logs.

  1. Run traceroute, ping, to capture pod logs.
  2. Record bandwidth and latency test results.
  3. Build and run a Java Database Connectivity (JDBC) client program that periodically logs SQL execution times from inside the pods and hosting nodes. 
 
3

Report

Submit findings to network admin or BMC Support.

  1. Include connection logs and network diagram in the report.
  2. To raise a support ticket, refer to Submit & Track Support Cases - BMC Documentation
BMC Helix Support
4Apply fix from Support
  • Apply the fix provided by BMC Helix Support.
 

Diagnosing and reporting application issues

Perform task 1 to troubleshoot the issue. If you are unable to resolve the issue, perform tasks 2 to 4 to create a BMC Helix Support ticket.

Task

Action

Steps

Reference

1

Troubleshoot and self-resolve

Compare the configurations and review the logs.

Adjust the logging and remove outdated data.

  1. Review and compare configuration files between the source (virtual machine–based) and target (containerized) environments. 
  2. Check the Centralized Automation Interface (CAI) plug-in execution logs, including API, filter, and SQL logs.  
  3. Ensure that BMC Helix Dashboard queries are routed through the platform-sr component and that they connect to read-only database instances. 
  4. Review and compare Application Request System (AR System) startup logs from both the source and target environments. 
  5. Set logging levels to minimal during business hours on production systems. 
  6. Run stale data cleanup jobs on high-transaction forms such as SRM:Variable and Historic_Process_Context, especially during onboarding or migration activities.
 
2

Collect diagnostics

Gather thread dumps and logs.

  1. Capture thread dumps from all relevant platform FTS pods, such as server and itsmplugin. 
  2. For delays during reporting, capture thread dumps from the Pentaho Carte server. 
 
3

Report

Escalate to Support with evidence.

  1. Include logs, thread dumps, and configuration diffs in the report.
  2. To raise a support ticket, refer to Submit & Track Support Cases - BMC Documentation
BMC Helix Support
4Apply fix from Support
  • Apply the fix provided by BMC Helix Support.
 

Best practices

Follow these best practices to maintain and improve performance in a containerized deployment: 

  • Automate statistics updates using the AUTO_UPDATE_STATISTICS feature in Microsoft SQL Server database. 
  • Monitor Long Running API (LRA) and Long Running Query (LRQ) reports regularly. 
  • Capture and analyze thread dumps at least once per week during peak load hours. 
  • Review database indexes and check for blocking queries as part of your regular maintenance schedule. 
  • Always validate configuration settings after migration to ensure consistency across environments. 

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

BMC Helix Service Management Deployment 25.3.01