March 2026 (2026-03) monthly summary for Linkerd2 development focusing on monitoring data quality and reliability. Key change: Monitoring Scraping: Exclude Non-Running Pods to Improve Data Quality. Updated the configmap so that only pods in running or pending state are scraped, thereby improving monitoring accuracy by excluding failed pods and completed jobs. This directly reduces noise in metrics and dashboards and supports more reliable alerting. Highlights: - Implemented config-driven filtering in monitoring scraping to exclude non-running pods (resolved issue #10562). Change references commit a913c05338fc1ae9bec326d2b4bd5a24e1373868 and related test updates (PR #14916). - Updated and expanded test fixtures to validate the new scraping behavior, increasing test coverage for data quality controls. - CI and code review completed with documentation updated for on-call dashboards and operator guidance. Business/value rationale: - Improves observability fidelity, enabling faster incident detection, more accurate SLA reporting, and reduced false alerts, empowering SRE and product teams to act with confidence. Technologies/skills demonstrated: - Kubernetes config management (configmaps), monitoring/observability practices, test fixture maintenance, PR review and collaboration across teams.