Change Management

Enhanced Definition

Change Management, in the mainframe context, refers to the structured process of controlling, tracking, and auditing modifications made to system software, application code, configurations, and data within the z/OS environment. Its primary purpose is to ensure the stability, integrity, and reliability of critical mainframe systems by minimizing risks associated with changes.

Key Characteristics

- Formalized Process: Involves predefined steps, approval workflows, and documentation requirements for every modification, from development to production.
- Version Control Integration: Often tightly coupled with Software Configuration Management (SCM) tools (e.g., CA Endevor SCM, IBM z/OS Change Tracker) to manage different versions of source code, JCL, and configuration files.
- Impact Analysis: Requires a thorough assessment of potential risks, dependencies, and resource implications before a change is approved and implemented.
- Backout/Recovery Plans: Mandates the creation and testing of procedures to revert changes quickly and safely if unforeseen issues arise post-implementation.
- Audit Trail and Compliance: Maintains a comprehensive history of all changes, including who made them, when, why, and their status, which is crucial for regulatory compliance (e.g., SOX, PCI DSS).
- Environment Promotion: Manages the controlled movement of changes through various environments (e.g., Development, Test, QA, Production LPARs).

Use Cases

- Application Development and Maintenance: Managing the lifecycle of changes to COBOL programs, JCL procedures, REXX scripts, DB2 DDL, or CICS definitions as they move from development to production.
- System Software Updates: Controlling the application of Program Temporary Fixes (PTFs), Authorized Program Analysis Reports (APARs), or new releases of z/OS, DB2, CICS, IMS, or other system utilities.
- Configuration Modifications: Implementing changes to critical system parameters in SYS1.PARMLIB, PROCLIB members, network definitions (VTAM), or security rules (RACF).
- Database Schema Changes: Managing alterations to DB2 tables, indexes, views, or IMS DBDs and PSBs, ensuring data integrity and application compatibility.
- Emergency Fixes: Providing a streamlined, yet controlled, process for implementing urgent fixes to critical production issues while maintaining an audit trail.

Related Concepts

Change Management is a foundational component of IT Service Management (ITSM) and is closely related to Software Configuration Management (SCM), which provides the tools and repositories for versioning and storing code. It works hand-in-hand with Release Management by ensuring all changes bundled into a release are properly controlled and tested, and with Problem Management where a problem often triggers a formal change request. It also underpins Disaster Recovery Planning by ensuring that all system configurations and application code are consistently managed and recoverable.

Best Practices:

Categorize Changes: Classify changes (e.g., Standard, Normal, Emergency) to apply appropriate levels of scrutiny, approval, and testing based on their potential impact and urgency.
Implement Robust Testing: Ensure changes are thoroughly tested in dedicated, isolated LPARs or environments (e.g., unit, integration, system, user acceptance testing) before deployment to production.
Automate Workflows: Leverage specialized mainframe change management tools to automate approval processes, deployments, and backout procedures, reducing manual errors and improving efficiency.
Comprehensive Documentation: Maintain clear, concise, and up-to-date documentation for every change, including its rationale, impact analysis, test results, and detailed backout instructions.
Regular Audits and Reviews: Conduct periodic audits of change records and processes to ensure compliance with organizational policies, regulatory requirements, and industry best practices.