Every engineer has stared at two dashboards and wondered why they do not talk to each other. New Relic tells you what is breaking. Phabricator tells you who broke it. Together they ought to form a closed loop of insight and accountability. Yet too often, that loop involves copy-paste heroics and stale alert comments.
New Relic is the observability oracle for code and infrastructure. Phabricator is the forge where reviews, tasks, and deployments live. One reveals data at runtime, the other organizes the humans responding to it. When integrated, incidents get context and commits get telemetry. You stop guessing which push turned your latency graph into modern art.
Here is how the New Relic–Phabricator pairing really works. Alerts flow from New Relic via webhook or custom API to Phabricator tasks. Those tasks link directly to affected services, revisions, and authors. Identity maps through the same provider you use for SSO: Okta, Google Workspace, or any OIDC-compliant IAM like AWS Cognito. The result is a traceable path from performance metric to decision. Every alert lands with provenance, not noise.
A clean setup depends on three small details. First, map service names consistently with your deployment pipeline so that tags in New Relic match repository paths in Phabricator. Second, enforce RBAC on the Phabricator side to keep sensitive metrics visible only to authorized users. Third, rotate API credentials often and store them in your secrets manager. These tweaks prevent stale keys and phantom alerts.
Quick answer: What does New Relic Phabricator integration actually do?
It links performance data to code reviews so teams can see which commits caused which issues, who handled them, and how fixes affected uptime—all inside one workflow without juggling dashboards.