fix: reduce parallelism and increase worker size on go-test-race (#15106)

spikecurtis · web-flow · commit d18e8304d6d1 · 2024-10-18T10:45:31.000+04:00
Sets parallelism on go-test-race to 4 concurrent tests and 4 concurrent
packages. Increases to 16-core runner.
diff --git a/.github/workflows/ci.yaml b/.github/workflows/ci.yaml
@@ -466,7 +466,7 @@ jobs:
           api-key: ${{ secrets.DATADOG_API_KEY }}
 
   test-go-race:
-    runs-on: ${{ github.repository_owner == 'coder' && 'depot-ubuntu-22.04-8' || 'ubuntu-latest' }}
+    runs-on: ${{ github.repository_owner == 'coder' && 'depot-ubuntu-22.04-16' || 'ubuntu-latest' }}
     needs: changes
     if: needs.changes.outputs.go == 'true' || needs.changes.outputs.ci == 'true' || github.ref == 'refs/heads/main'
     timeout-minutes: 25
@@ -487,9 +487,13 @@ jobs:
       - name: Setup Terraform
         uses: ./.github/actions/setup-tf
 
+      # We run race tests with reduced parallelism because they use more CPU and we were finding
+      # instances where tests appear to hang for multiple seconds, resulting in flaky tests when
+      # short timeouts are used.
+      # c.f. discussion on https://github.com/coder/coder/pull/15106
       - name: Run Tests
         run: |
-          gotestsum --junitfile="gotests.xml" -- -race ./...
+          gotestsum --junitfile="gotests.xml" -- -race -parallel 4 -p 4 ./...
 
       - name: Upload test stats to Datadog
         timeout-minutes: 1
diff --git a/Makefile b/Makefile
@@ -817,7 +817,7 @@ test-postgres-docker:
 
 # Make sure to keep this in sync with test-go-race from .github/workflows/ci.yaml.
 test-race:
-	$(GIT_FLAGS) gotestsum --junitfile="gotests.xml" -- -race -count=1 ./...
+	$(GIT_FLAGS) gotestsum --junitfile="gotests.xml" -- -race -count=1 -parallel 4 -p 4 ./...
 .PHONY: test-race
 
 test-tailnet-integration:
