From 460d8e468a73e2d8edee625b9a9fe97ef24fa36f Mon Sep 17 00:00:00 2001
From: Mike Sarahan <msarahan@gmail.com>
Date: Fri, 22 Nov 2024 16:36:57 -0600
Subject: [PATCH] add telemetry (#4740)

Enables telemetry during cugraph's build process. This parses github job metadata to obtain timing information. It should have very little impact on overall build time, and should not interfere with any build tools.

This implements emitting OpenTelemetry traces and spans, as described in https://github.com/rapidsai/build-infra/issues/139

Authors:
  - Mike Sarahan (https://github.com/msarahan)

Approvers:
  - Bradley Dice (https://github.com/bdice)

URL: https://github.com/rapidsai/cugraph/pull/4740
---
 .github/workflows/pr.yaml | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml
index e82342dfd9..c8bf94b098 100644
--- a/.github/workflows/pr.yaml
+++ b/.github/workflows/pr.yaml
@@ -10,6 +10,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
+  # Please keep pr-builder as the top job here
   pr-builder:
     needs:
       - changed-files
@@ -25,14 +26,24 @@ jobs:
       - wheel-tests-pylibcugraph
       - wheel-build-cugraph
       - wheel-tests-cugraph
+      - telemetry-setup
       - devcontainer
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/pr-builder.yaml@branch-24.12
     if: always()
     with:
       needs: ${{ toJSON(needs) }}
+  telemetry-setup:
+    runs-on: ubuntu-latest
+    continue-on-error: true
+    env:
+        OTEL_SERVICE_NAME: "pr-cugraph"
+    steps:
+      - name: Telemetry setup
+        uses: rapidsai/shared-actions/telemetry-dispatch-stash-base-env-vars@main
   changed-files:
     secrets: inherit
+    needs: telemetry-setup
     uses: rapidsai/shared-workflows/.github/workflows/changed-files.yaml@branch-24.12
     with:
       files_yaml: |
@@ -63,9 +74,11 @@ jobs:
           - '!notebooks/**'
   checks:
     secrets: inherit
+    needs: telemetry-setup
     uses: rapidsai/shared-workflows/.github/workflows/checks.yaml@branch-24.12
     with:
       enable_check_generated_files: false
+      ignored_pr_jobs: telemetry-summarize
   conda-cpp-build:
     needs: checks
     secrets: inherit
@@ -161,6 +174,7 @@ jobs:
       script: ci/test_wheel_cugraph.sh
   devcontainer:
     secrets: inherit
+    needs: telemetry-setup
     uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-24.12
     with:
       arch: '["amd64"]'
@@ -171,3 +185,17 @@ jobs:
         sccache -z;
         build-all --verbose -j$(nproc --ignore=1) -DBUILD_CUGRAPH_MG_TESTS=ON;
         sccache -s;
+  telemetry-summarize:
+    runs-on: ubuntu-latest
+    needs: pr-builder
+    if: always()
+    continue-on-error: true
+    steps:
+      - name: Load stashed telemetry env vars
+        uses: rapidsai/shared-actions/telemetry-dispatch-load-base-env-vars@main
+        with:
+            load_service_name: true
+      - name: Telemetry summarize
+        uses: rapidsai/shared-actions/telemetry-dispatch-write-summary@main
+        with:
+          cert_concat: "${{ secrets.OTEL_EXPORTER_OTLP_CA_CERTIFICATE }};${{ secrets.OTEL_EXPORTER_OTLP_CLIENT_CERTIFICATE }};${{ secrets.OTEL_EXPORTER_OTLP_CLIENT_KEY }}"