Fix E2E job: Implement robust server startup with proper health checks and cleanup
CI Pipeline / canary (pull_request) Successful in 1s
CI Pipeline / test (20) (pull_request) Successful in 1m50s
CI Pipeline / test (18) (pull_request) Successful in 2m7s
CI Pipeline / e2e (chromium) (pull_request) Has been cancelled
CI Pipeline / e2e (webkit) (pull_request) Has been cancelled
CI Pipeline / visual-regression (pull_request) Has been cancelled
CI Pipeline / performance (pull_request) Has been cancelled
CI Pipeline / storybook (pull_request) Has been cancelled
CI Pipeline / lint (pull_request) Has been cancelled
CI Pipeline / build (pull_request) Has been cancelled
CI Pipeline / e2e (firefox) (pull_request) Has been cancelled

This commit is contained in:
adilallo
2025-08-29 23:28:28 -06:00
parent 7b660e41eb
commit 92f67dc678
+37 -56
View File
@@ -60,66 +60,40 @@ jobs:
- run: npx playwright install --with-deps ${{ matrix.browser }}
- run: npm run build
# 1) Sanity check that the build exists
- name: Verify Next build output
- name: Start app (background) + healthcheck
run: |
set -euxo pipefail
ls -la .next || true
test -f .next/BUILD_ID || (echo "No Next build output (.next) did build fail?" && exit 1)
# pick a port that's unlikely to be busy
export PORT="${PORT:-3010}"
export HOST="127.0.0.1"
# 2) Start the right kind of server and capture logs
- name: Start app (auto-detect start vs export)
# ensure build exists
test -d .next || { echo "❌ Missing .next build output"; exit 1; }
# start and detach with logs
mkdir -p .next
nohup npm run start -- -p "$PORT" -H "$HOST" > .next/runner.log 2>&1 &
echo $! > .next/runner.pid
echo "🌐 PID $(cat .next/runner.pid) listening on http://$HOST:$PORT"
# wait for TCP, then HTTP
npx wait-on -t 120000 "tcp:$HOST:$PORT"
curl -fsS "http://$HOST:$PORT" >/dev/null
echo "✅ App is responding"
env:
NEXT_TELEMETRY_DISABLED: "1"
NODE_ENV: production
- name: Show last 200 lines of server log on failure
if: failure()
run: |
set -euxo pipefail
# prefer production server if Next build exists
if [ -f .next/BUILD_ID ]; then
# run Next in the foreground for 3s to flush any crash to logs, then background it
(NODE_ENV=production NEXT_TELEMETRY_DISABLED=1 \
node --trace-uncaught node_modules/next/dist/bin/next start -p 3000 -H 127.0.0.1 \
) > server.log 2>&1 &
# fall back to static export (if your project uses output: 'export')
elif [ -d out ]; then
npx --yes http-server out -p 3000 --silent > server.log 2>&1 &
else
echo "Neither .next nor out/ present nothing to serve"; cat server.log || true; exit 1
fi
echo $! > server.pid
sleep 2
echo "----- first server log lines -----"
head -n 200 server.log || true
echo "----------------------------------"
# 3) Fail fast if the process died
- name: Check server process
run: |
set -e
PID=$(cat server.pid)
if ! ps -p "$PID" > /dev/null; then
echo "Server crashed during startup:"
cat server.log || true
exit 1
fi
# 4) Wait for readiness with helpful diagnostics
- name: Wait for http://127.0.0.1:3000
run: |
set -euxo pipefail
for i in $(seq 1 60); do
if curl -sf http://127.0.0.1:3000 >/dev/null; then
echo "App is up ✅"; exit 0
fi
if [ $i -eq 1 ] || [ $((i%10)) -eq 0 ]; then
echo "Still waiting… attempt $i"
tail -n 50 server.log || true
fi
sleep 2
done
echo "Timed out waiting for app"; tail -n +1 server.log || true; exit 1
- run: npx playwright test --project=${{ matrix.browser }}
env: { CI: true }
echo "––– .next/runner.log (tail) –––"
tail -n 200 .next/runner.log || true
- name: Run E2E tests
run: npx playwright test --project=${{ matrix.browser }}
env:
CI: true
BASE_URL: http://127.0.0.1:3010
# package artifacts (keeps file count small)
- name: Package E2E artifacts
@@ -134,6 +108,13 @@ jobs:
name: playwright-results-${{ matrix.browser }}
path: playwright-${{ matrix.browser }}.tgz
retention-days: 30
- name: Stop app
if: always()
run: |
if [ -f .next/runner.pid ]; then
kill $(cat .next/runner.pid) 2>/dev/null || true
fi
visual-regression:
runs-on: [self-hosted, macos-latest]