This community innovation has been accepted at the 2026 DHIS2 Annual Conference and will be in abstract track/lightning talk.
Metadata as Code: Versioning DHIS2 Configurations
DHIS2 is a highly configurable, purpose-agnostic platform whose behavior is almost entirely defined through metadata. By modifying metadata alone, a DHIS2 instance can support logistics management, education systems, civil registries, or patient-level surveillance. While this flexibility is one of DHIS2’s greatest strengths, it also introduces significant complexity in managing, validating, and governing metadata over time. In many DHIS2 implementations, substantial attention is given to data quality, traceability, and auditability. However, data cannot be correctly interpreted without knowing the exact metadata configuration under which it was collected. Despite this, metadata development processes are often manual, opaque, and weakly versioned. Teams frequently lack mechanisms to answer fundamental questions: How has metadata evolved over time? Can we reproduce a past system state? How do multiple developers collaborate safely on metadata? How can changes be reviewed, validated, merged, and rolled back? Drawing from EyeSeeTea’s experience as a DHIS2-focused development team working across large, multi-country implementations, this work introduces a “metadata-as-code” approach to DHIS2 metadata management. We present a set of practices and tools that treat metadata similarly to software source code, enabling structured versioning, change tracking, automated validation, quality assurance, and collaborative workflows. The approach supports comparing metadata states, managing parallel development efforts, merging metadata packages, and ensuring reproducibility across environments. By aligning metadata development with established software engineering practices, this work contributes a practical, scalable strategy for improving governance, transparency, and long-term sustainability of DHIS2 implementations.
Primary Author: Ignacio Foche Pérez
Keywords:
DHIS2, metadata management, version control, metadata-as-code, quality assurance, governance, DevOps
2 Likes