test reports for test.sh and fix test.sh deployments (#6027)

.github/workflows/build.yml

@@ -389,6 +389,7 @@ jobs:
           MAVEN_USER: ${{ secrets.MAVEN_USER }}
           MAVEN_PASSWORD: ${{ secrets.MAVEN_PASSWORD }}
           MAVEN_PUBLIC_URL: ${{ secrets.MAVEN_PUBLIC_URL }}
+          DOCKER_BASE_CHANGED: ${{ needs.files-changed.outputs.docker-base }}
 
       - name: Upload Cucumber Report
         if: always()
@@ -399,6 +400,15 @@ jobs:
           retention-days: 7
           if-no-files-found: warn
 
+      - name: Upload Test Reports
+        if: always()
+        uses: actions/upload-artifact@bbbca2ddaa5d8feaa63e36b76fdaad77386f024f # v7.0.0
+        with:
+          name: docker-compose-test-reports
+          path: testing/reports/
+          retention-days: 7
+          if-no-files-found: warn
+
       - name: Cucumber Test Report
         if: always()
         uses: dorny/test-reporter@b082adf0eced0765477756c2a610396589b8c637 # v2.5.0

build.gradle

@@ -177,8 +177,6 @@ subprojects {
         exclude group: 'org.bouncycastle', module: 'bcpkix-jdk15on'
         exclude group: 'org.bouncycastle', module: 'bcutil-jdk15on'
         exclude group: 'org.bouncycastle', module: 'bcmail-jdk15on'
-        // Force a compatible Guava version for spotless
-        resolutionStrategy.force 'com.google.guava:guava:33.0.0-jre'
 
         // Security CVE fixes - hardcoded resolution strategy to ensure safe versions
         // Primary fixes via explicit dependencies in app/core/build.gradle:

docker/embedded/compose/docker-compose-latest-fat-endpoints-disabled.yml

@@ -2,9 +2,6 @@
 services:
   stirling-pdf:
     container_name: Stirling-PDF-Fat-Disable-Endpoints
-    build:
-      context: ../../..
-      dockerfile: docker/embedded/Dockerfile.fat
     image: docker.stirlingpdf.com/stirlingtools/stirling-pdf:fat
     build:
       context: ../../../

docker/embedded/compose/docker-compose-latest-fat-security.yml

@@ -1,13 +1,10 @@
 services:
   stirling-pdf:
     container_name: Stirling-PDF-Security-Fat
-    build:
-      context: ../../..
-      dockerfile: docker/embedded/Dockerfile.fat
     image: docker.stirlingpdf.com/stirlingtools/stirling-pdf:fat
     build:
       context: ../../../
-      dockerfile: docker/embedded/Dockerfile
+      dockerfile: docker/embedded/Dockerfile.fat
     deploy:
       resources:
         limits:

docker/embedded/compose/docker-compose-latest-ultra-lite.yml

@@ -1,9 +1,6 @@
 services:
   stirling-pdf:
     container_name: Stirling-PDF-Ultra-Lite
-    build:
-      context: ../../..
-      dockerfile: docker/embedded/Dockerfile.ultra-lite
     image: docker.stirlingpdf.com/stirlingtools/stirling-pdf:ultra-lite
     build:
       context: ../../../

docker/embedded/compose/test_cicd.yml

@@ -1,9 +1,6 @@
 services:
   stirling-pdf:
     container_name: Stirling-PDF-Security-Fat-with-login
-    build:
-      context: ../../..
-      dockerfile: docker/embedded/Dockerfile.fat
     image: docker.stirlingpdf.com/stirlingtools/stirling-pdf:fat
     build:
       context: ../../../

testing/cucumber/features/environment.py

@@ -1,8 +1,11 @@
 import os
+import subprocess
 
 import requests
 
 _BASE_URL = "http://localhost:8080"
+_CONTAINER_NAME = os.environ.get("TEST_CONTAINER_NAME", "")
+_REPORT_DIR = os.environ.get("TEST_REPORT_DIR", "")
 
 # Tags that indicate a scenario requires JWT Bearer auth to be functional.
 # These scenarios are skipped when the server has JWT disabled (V2=false).
@@ -44,6 +47,44 @@ def _check_jwt_available():
         return False
 
 
+def _get_docker_log_line_count():
+    if not _CONTAINER_NAME:
+        return 0
+    try:
+        result = subprocess.run(
+            ["docker", "logs", _CONTAINER_NAME],
+            capture_output=True, text=True, timeout=10,
+        )
+        return len(result.stdout.splitlines()) + len(result.stderr.splitlines())
+    except Exception:
+        return 0
+
+
+def _capture_docker_logs_window(start_line, scenario_name):
+    if not _CONTAINER_NAME or not _REPORT_DIR:
+        return
+    try:
+        result = subprocess.run(
+            ["docker", "logs", _CONTAINER_NAME],
+            capture_output=True, text=True, timeout=10,
+        )
+        all_lines = (result.stdout + result.stderr).splitlines()
+        window = all_lines[start_line:]
+        if not window:
+            return
+
+        safe_name = "".join(c if c.isalnum() or c in "-_" else "_" for c in scenario_name)
+        log_path = os.path.join(_REPORT_DIR, f"scenario_{safe_name}.log")
+        with open(log_path, "w") as f:
+            f.write(f"=== Docker logs during: {scenario_name} ===\n")
+            f.write(f"Container: {_CONTAINER_NAME}\n")
+            f.write(f"Lines {start_line + 1}-{len(all_lines)} ({len(window)} lines)\n")
+            f.write("---\n")
+            f.write("\n".join(window[-200:]) + "\n")
+    except Exception:
+        pass
+
+
 def before_all(context):
     context.endpoint = None
     context.request_data = None
@@ -81,8 +122,14 @@ def before_scenario(context, scenario):
     # Stored value used by enterprise step definitions for dynamic URL composition
     context.stored_value = None
 
+    context._docker_log_line_count = _get_docker_log_line_count()
+
 
 def after_scenario(context, scenario):
+    if scenario.status == "failed":
+        start_line = getattr(context, "_docker_log_line_count", 0)
+        _capture_docker_logs_window(start_line, scenario.name)
+
     if hasattr(context, "files"):
         for file in context.files.values():
             try:

testing/test.sh

@@ -31,45 +31,221 @@ find_root() {
 
 PROJECT_ROOT=$(find_root)
 
-# Base image version - must be provided or read from environment
+REPORT_DIR="$PROJECT_ROOT/testing/reports"
-# This is a testing-specific version; production should pass explicit BASE_VERSION
+mkdir -p "$REPORT_DIR"
-if [ -z "$BASE_VERSION" ]; then
+
-    # For CI/automation: use a unique test identifier
+declare -A test_start_times
-    if [ -n "${GITHUB_RUN_ID}" ]; then
+declare -A test_durations
-        BASE_VERSION="test-${GITHUB_RUN_ID}"
+declare -A test_failure_logs
+CURRENT_CONTAINER=""
+
+is_gha() {
+    [ -n "${GITHUB_ACTIONS:-}" ]
+}
+
+gha_group() {
+    if is_gha; then echo "::group::$1"; else echo "=== $1 ==="; fi
+}
+gha_endgroup() {
+    if is_gha; then echo "::endgroup::"; fi
+}
+
+start_test_timer() {
+    local test_name=$1
+    test_start_times["$test_name"]=$SECONDS
+}
+
+stop_test_timer() {
+    local test_name=$1
+    local start=${test_start_times["$test_name"]:-$SECONDS}
+    test_durations["$test_name"]=$(( SECONDS - start ))
+}
+
+capture_failure_logs() {
+    local test_name=$1
+    local container_name=$2
+    local extra_context=$3
+    local log_file="$REPORT_DIR/${test_name//[^a-zA-Z0-9_-]/_}.failure.log"
+
+    {
+        echo "=== Failure logs for: $test_name ==="
+        echo "Timestamp: $(date -u +%Y-%m-%dT%H:%M:%SZ)"
+        if [ -n "$container_name" ]; then
+            echo "Container: $container_name"
+            echo "---"
+            docker logs "$container_name" 2>&1 | tail -200
+        elif [ -n "$extra_context" ]; then
+            echo "---"
+            echo "$extra_context"
         else
-        # For local testing: generate unique identifier
+            echo "---"
-        BASE_VERSION="test-local-$(date +%s)"
+            echo "No container was running. Docker-compose may have failed to start."
+        fi
+    } > "$log_file" 2>/dev/null || true
+    test_failure_logs["$test_name"]="$log_file"
+}
+
+capture_build_failure() {
+    local build_name=$1
+    local log_file="$REPORT_DIR/${build_name//[^a-zA-Z0-9_-]/_}.failure.log"
+    local gradle_report_dirs=(
+        "$PROJECT_ROOT/app/core/build/reports/tests"
+        "$PROJECT_ROOT/app/common/build/reports/tests"
+        "$PROJECT_ROOT/app/proprietary/build/reports/tests"
+    )
+
+    {
+        echo "=== Build failure: $build_name ==="
+        echo "Timestamp: $(date -u +%Y-%m-%dT%H:%M:%SZ)"
+        echo "---"
+
+        for report_dir in "${gradle_report_dirs[@]}"; do
+            if [ -d "$report_dir" ]; then
+                local txt_index="$report_dir/test/index.html"
+                if [ -f "$txt_index" ]; then
+                    echo "--- Gradle test report: $report_dir ---"
+                    sed 's/<[^>]*>//g' "$txt_index" | head -100
+                    echo ""
                 fi
             fi
-BASE_IMAGE="ghcr.io/stirling-tools/stirling-pdf-base:${BASE_VERSION}"
+        done
 
-# Function to ensure base image exists (build if missing)
+        for xml_file in "$PROJECT_ROOT"/app/*/build/test-results/test/TEST-*.xml; do
-ensure_base_image() {
+            if [ -f "$xml_file" ]; then
-    echo "Checking for base image: $BASE_IMAGE"
+                local failures
-
+                failures=$(grep -c 'failures="[^0]' "$xml_file" 2>/dev/null || true)
-    if docker image inspect "$BASE_IMAGE" >/dev/null 2>&1; then
+                local errors
-        echo "✓ Base image found locally: $BASE_IMAGE"
+                errors=$(grep -c 'errors="[^0]' "$xml_file" 2>/dev/null || true)
-        return 0
+                if [ "$failures" -gt 0 ] || [ "$errors" -gt 0 ]; then
+                    echo "--- Failed: $(basename "$xml_file") ---"
+                    grep -A 5 '<failure\|<error' "$xml_file" | head -50
+                    echo ""
                 fi
-
-    echo "Base image not found. Attempting to pull from registry..."
-    if docker pull "$BASE_IMAGE" 2>/dev/null; then
-        echo "✓ Pulled base image from registry: $BASE_IMAGE"
-        return 0
             fi
+        done
+    } > "$log_file" 2>/dev/null || true
 
-    echo "Base image not available in registry. Building from source..."
+    test_failure_logs["$build_name"]="$log_file"
+}
+
+
+generate_json_report() {
+    local report_file="$REPORT_DIR/test-report.json"
+    local total_duration=$SECONDS
+    local total_tests=$(( ${#passed_tests[@]} + ${#failed_tests[@]} ))
+
+    {
+        echo "{"
+        echo "  \"summary\": {"
+        echo "    \"total\": $total_tests,"
+        echo "    \"passed\": ${#passed_tests[@]},"
+        echo "    \"failed\": ${#failed_tests[@]},"
+        echo "    \"duration_seconds\": $total_duration,"
+        echo "    \"result\": \"$([ ${#failed_tests[@]} -eq 0 ] && echo 'PASS' || echo 'FAIL')\","
+        echo "    \"timestamp\": \"$(date -u +%Y-%m-%dT%H:%M:%SZ)\""
+        echo "  },"
+        echo "  \"tests\": ["
+
+        local first=true
+        for test in "${passed_tests[@]}"; do
+            if [ "$first" = true ]; then first=false; else echo ","; fi
+            local dur=${test_durations["$test"]:-0}
+            printf "    {\"name\": \"%s\", \"status\": \"PASS\", \"duration_seconds\": %d}" "$test" "$dur"
+        done
+
+        for test in "${failed_tests[@]}"; do
+            if [ "$first" = true ]; then first=false; else echo ","; fi
+            local dur=${test_durations["$test"]:-0}
+            local log_file=${test_failure_logs["$test"]:-""}
+            local failure_snippet=""
+            if [ -n "$log_file" ] && [ -f "$log_file" ]; then
+                # Grab last 20 lines as a failure snippet, escape for JSON
+                failure_snippet=$(tail -20 "$log_file" | sed 's/\\/\\\\/g; s/"/\\"/g; s/\t/\\t/g' | tr '\n' '|' | sed 's/|/\\n/g')
+            fi
+            printf "    {\"name\": \"%s\", \"status\": \"FAIL\", \"duration_seconds\": %d, \"log_file\": \"%s\", \"failure_snippet\": \"%s\"}" \
+                "$test" "$dur" "$log_file" "$failure_snippet"
+        done
+
+        echo ""
+        echo "  ]"
+        echo "}"
+    } > "$report_file"
+
+    echo "JSON test report written to: $report_file"
+}
+
+generate_gha_summary() {
+    if ! is_gha; then return; fi
+    local summary_file="${GITHUB_STEP_SUMMARY}"
+    local total_duration=$SECONDS
+
+    {
+        echo "## Docker Compose Test Results"
+        echo ""
+        if [ ${#failed_tests[@]} -eq 0 ]; then
+            echo "**Result: ALL PASSED** in ${total_duration}s"
+        else
+            echo "**Result: ${#failed_tests[@]} FAILED** out of $(( ${#passed_tests[@]} + ${#failed_tests[@]} )) tests in ${total_duration}s"
+        fi
+        echo ""
+        echo "| Test | Status | Duration |"
+        echo "|------|--------|----------|"
+
+        for test in "${passed_tests[@]}"; do
+            local dur=${test_durations["$test"]:-0}
+            echo "| ${test} | :white_check_mark: PASS | ${dur}s |"
+        done
+
+        for test in "${failed_tests[@]}"; do
+            local dur=${test_durations["$test"]:-0}
+            echo "| ${test} | :x: FAIL | ${dur}s |"
+        done
+
+        # If there are failures, include snippets
+        if [ ${#failed_tests[@]} -ne 0 ]; then
+            echo ""
+            echo "### Failure Details"
+            echo ""
+            for test in "${failed_tests[@]}"; do
+                local log_file=${test_failure_logs["$test"]:-""}
+                echo "<details>"
+                echo "<summary>${test}</summary>"
+                echo ""
+                if [ -n "$log_file" ] && [ -f "$log_file" ]; then
+                    echo '```'
+                    tail -50 "$log_file"
+                    echo '```'
+                else
+                    echo "No failure logs captured. Check the full build log."
+                fi
+                echo "</details>"
+                echo ""
+            done
+        fi
+    } >> "$summary_file"
+
+    echo "GitHub Actions job summary written."
+}
+
+prepare_base_image() {
+    if [ "${DOCKER_BASE_CHANGED:-false}" = "true" ]; then
+        echo "Docker base files changed — building base image locally..."
+        gha_group "Build: Base image (local)"
         if docker build -f "$PROJECT_ROOT/docker/base/Dockerfile" \
-        -t "$BASE_IMAGE" \
+            -t stirling-pdf-base:local \
-        --build-arg BASE_VERSION="$BASE_VERSION" \
             "$PROJECT_ROOT/docker/base"; then
-        echo "✓ Built base image: $BASE_IMAGE"
+            echo "✓ Built base image locally: stirling-pdf-base:local"
-        return 0
+            BASE_IMAGE_ARG="--build-arg BASE_IMAGE=stirling-pdf-base:local"
         else
             echo "ERROR: Failed to build base image"
+            gha_endgroup
             return 1
         fi
+        gha_endgroup
+    else
+        echo "Docker base unchanged — using published base image from Dockerfile defaults"
+        BASE_IMAGE_ARG=""
+    fi
 }
 
 # Function to check application readiness via HTTP instead of Docker's health status
@@ -87,6 +263,7 @@ check_health() {
         echo "Using API key for health check: ${api_key:0:3}***"
     fi
 
+    gha_group "Health check: $container_name"
     echo "Waiting for $container_name to become reachable on http://localhost:8080/api/v1/info/status (timeout ${timeout}s)..."
     while [ $SECONDS -lt $end ]; do
         # Optional: check if container is running at all (nice for debugging)
@@ -104,8 +281,10 @@ check_health() {
         # Treat any 2xx as "ready"
         if [ "$last_code" -ge 200 ] && [ "$last_code" -lt 300 ]; then
             echo "$container_name is reachable over HTTP (status $last_code)."
-            echo "Printing logs for $container_name:"
+            gha_group "Container logs (startup): $container_name"
             docker logs "$container_name" || true
+            gha_endgroup
+            gha_endgroup
             return 0
         fi
 
@@ -120,8 +299,10 @@ check_health() {
     docker_health=$(docker inspect --format='{{if .State.Health}}{{.State.Health.Status}}{{else}}(no healthcheck){{end}}' "$container_name" 2>/dev/null || echo "inspect failed")
     echo "Docker-reported health status for $container_name: $docker_health"
 
-    echo "Printing logs for $container_name:"
+    gha_group "Container logs (failure): $container_name"
     docker logs "$container_name" || true
+    gha_endgroup
+    gha_endgroup
     return 1
 }
 
@@ -130,6 +311,7 @@ capture_file_list() {
     local container_name=$1
     local output_file=$2
 
+    gha_group "Capture file list: $container_name"
     echo "Capturing file list from $container_name..."
     # Get all files in one command, output directly from Docker to avoid path issues
     # Skip proc, sys, dev, and the specified LibreOffice config directory
@@ -194,6 +376,7 @@ capture_file_list() {
     fi
 
     echo "File list captured to $output_file"
+    gha_endgroup
 }
 
 # Function to compare before and after file lists
@@ -203,6 +386,7 @@ compare_file_lists() {
     local diff_file=$3
     local container_name=$4  # Added container_name parameter
 
+    gha_group "Compare file lists"
     echo "Comparing file lists..."
 
     # Check if files exist and have content
@@ -223,11 +407,13 @@ compare_file_lists() {
                 cat "${diff_file}.tmp"
                 echo "Printing docker logs due to temporary file detection:"
                 docker logs "$container_name"  # Print logs when temp files are found
+                gha_endgroup
                 return 1
             else
                 echo "No temporary files found in the after snapshot."
             fi
         fi
+        gha_endgroup
         return 0
     fi
 
@@ -264,6 +450,7 @@ compare_file_lists() {
     else
         echo "No file changes detected during test."
     fi
+    gha_endgroup
     return 0
 }
 
@@ -313,6 +500,7 @@ test_compose() {
     local test_name=$2
     local status=0
 
+    gha_group "Deploy: $test_name"
     echo "Testing ${compose_file} configuration..."
 
     # Start up the Docker Compose service
@@ -326,10 +514,16 @@ test_compose() {
 
     if [[ -z "$container_name" ]]; then
         echo "ERROR: No running container found for ${compose_file}"
-        docker-compose -f "$compose_file" ps
+        local compose_output
+        compose_output=$(docker-compose -f "$compose_file" ps 2>&1)
+        echo "$compose_output"
+        capture_failure_logs "$test_name" "" "docker-compose failed for: ${compose_file}
+${compose_output}"
+        gha_endgroup
         return 1
     fi
 
+    CURRENT_CONTAINER="$container_name"
     echo "Started container: $container_name"
 
     # Wait for the service to become healthy (HTTP-based)
@@ -337,9 +531,11 @@ test_compose() {
         echo "${test_name} test passed."
     else
         echo "${test_name} test failed."
+        capture_failure_logs "$test_name" "$container_name"
         status=1
     fi
 
+    gha_endgroup
     return $status
 }
 
@@ -407,11 +603,19 @@ run_tests() {
         return 0
     fi
 
+    start_test_timer "$test_name"
     if test_compose "$compose_file" "$test_name"; then
         passed_tests+=("$test_name")
     else
         failed_tests+=("$test_name")
     fi
+    stop_test_timer "$test_name"
+}
+
+finalize_reports() {
+    generate_json_report
+    generate_gha_summary
+    save_failed_tests
 }
 
 # Main testing routine
@@ -419,11 +623,12 @@ main() {
     SECONDS=0
     cd "$PROJECT_ROOT"
 
-    # Ensure base image exists before running tests
+    trap finalize_reports EXIT
+
     echo "=========================================="
     echo "Preparing Docker base image..."
     echo "=========================================="
-    ensure_base_image || exit 1
+    prepare_base_image || exit 1
     echo ""
 
     # Parse command line arguments
@@ -472,9 +677,13 @@ main() {
        should_run_test "Webpage-Accessibility-lite" || \
        should_run_test "Stirling-PDF-Ultra-Lite-Version-Check"; then
 
+        gha_group "Build: Ultra-Lite (Gradle + Docker)"
         export DISABLE_ADDITIONAL_FEATURES=true
         if ! ./gradlew clean build -PnoSpotless; then
             echo "Gradle build failed with security disabled, exiting script."
+            failed_tests+=("Build-Ultra-Lite-Gradle")
+            capture_build_failure "Build-Ultra-Lite-Gradle"
+            gha_endgroup
             exit 1
         fi
 
@@ -490,11 +699,16 @@ main() {
         else
             DOCKER_CACHE_ARGS_ULTRA_LITE=""
         fi
-        docker buildx build --build-arg VERSION_TAG=alpha \
+        if ! docker buildx build --build-arg VERSION_TAG=alpha \
             -t docker.stirlingpdf.com/stirlingtools/stirling-pdf:ultra-lite \
             -f ./docker/embedded/Dockerfile.ultra-lite \
             --load \
-            ${DOCKER_CACHE_ARGS_ULTRA_LITE} .
+            ${DOCKER_CACHE_ARGS_ULTRA_LITE} .; then
+            failed_tests+=("Build-Ultra-Lite-Docker")
+            gha_endgroup
+            exit 1
+        fi
+        gha_endgroup
     else
         echo "Skipping ultra-lite image build - no ultra-lite tests in rerun list"
     fi
@@ -503,26 +717,34 @@ main() {
     run_tests "Stirling-PDF-Ultra-Lite" "./docker/embedded/compose/docker-compose-latest-ultra-lite.yml"
 
     if should_run_test "Webpage-Accessibility-lite"; then
+        start_test_timer "Webpage-Accessibility-lite"
+        gha_group "Test: Webpage-Accessibility-lite"
         echo "Testing webpage accessibility..."
         cd "testing"
         if ./test_webpages.sh -f webpage_urls.txt -b http://localhost:8080; then
             passed_tests+=("Webpage-Accessibility-lite")
         else
             failed_tests+=("Webpage-Accessibility-lite")
+            capture_failure_logs "Webpage-Accessibility-lite" "$CURRENT_CONTAINER"
             echo "Webpage accessibility lite tests failed"
         fi
         cd "$PROJECT_ROOT"
+        gha_endgroup
+        stop_test_timer "Webpage-Accessibility-lite"
     fi
 
     if should_run_test "Stirling-PDF-Ultra-Lite-Version-Check"; then
+        start_test_timer "Stirling-PDF-Ultra-Lite-Version-Check"
         echo "Testing version verification..."
         if verify_app_version "Stirling-PDF-Ultra-Lite" "http://localhost:8080"; then
             passed_tests+=("Stirling-PDF-Ultra-Lite-Version-Check")
             echo "Version verification passed for Stirling-PDF-Ultra-Lite"
         else
             failed_tests+=("Stirling-PDF-Ultra-Lite-Version-Check")
+            capture_failure_logs "Stirling-PDF-Ultra-Lite-Version-Check" "$CURRENT_CONTAINER"
             echo "Version verification failed for Stirling-PDF-Ultra-Lite"
         fi
+        stop_test_timer "Stirling-PDF-Ultra-Lite-Version-Check"
     fi
 
     docker-compose -f "./docker/embedded/compose/docker-compose-latest-ultra-lite.yml" down -v
@@ -540,9 +762,13 @@ main() {
        should_run_test "Disabled-Endpoints" || \
        should_run_test "Stirling-PDF-Fat-Disable-Endpoints-Version-Check"; then
 
+        gha_group "Build: Fat + Security (Gradle + Docker)"
         export DISABLE_ADDITIONAL_FEATURES=false
         if ! ./gradlew clean build -PnoSpotless; then
             echo "Gradle build failed with security enabled, exiting script."
+            failed_tests+=("Build-Fat-Gradle")
+            capture_build_failure "Build-Fat-Gradle"
+            gha_endgroup
             exit 1
         fi
 
@@ -557,11 +783,17 @@ main() {
         else
             DOCKER_CACHE_ARGS_FAT=""
         fi
-        docker buildx build --build-arg VERSION_TAG=alpha \
+        if ! docker buildx build --build-arg VERSION_TAG=alpha \
+            ${BASE_IMAGE_ARG} \
             -t docker.stirlingpdf.com/stirlingtools/stirling-pdf:fat \
             -f ./docker/embedded/Dockerfile.fat \
             --load \
-            ${DOCKER_CACHE_ARGS_FAT} .
+            ${DOCKER_CACHE_ARGS_FAT} .; then
+            failed_tests+=("Build-Fat-Docker")
+            gha_endgroup
+            exit 1
+        fi
+        gha_endgroup
     else
         echo "Skipping fat image build - no fat tests in rerun list"
     fi
@@ -570,26 +802,34 @@ main() {
     run_tests "Stirling-PDF-Security-Fat" "./docker/embedded/compose/docker-compose-latest-fat-security.yml"
 
     if should_run_test "Webpage-Accessibility-full"; then
+        start_test_timer "Webpage-Accessibility-full"
+        gha_group "Test: Webpage-Accessibility-full"
         echo "Testing webpage accessibility..."
         cd "testing"
         if ./test_webpages.sh -f webpage_urls_full.txt -b http://localhost:8080; then
             passed_tests+=("Webpage-Accessibility-full")
         else
             failed_tests+=("Webpage-Accessibility-full")
+            capture_failure_logs "Webpage-Accessibility-full" "$CURRENT_CONTAINER"
             echo "Webpage accessibility full tests failed"
         fi
         cd "$PROJECT_ROOT"
+        gha_endgroup
+        stop_test_timer "Webpage-Accessibility-full"
     fi
 
     if should_run_test "Stirling-PDF-Security-Fat-Version-Check"; then
+        start_test_timer "Stirling-PDF-Security-Fat-Version-Check"
         echo "Testing version verification..."
         if verify_app_version "Stirling-PDF-Security-Fat" "http://localhost:8080"; then
             passed_tests+=("Stirling-PDF-Security-Fat-Version-Check")
             echo "Version verification passed for Stirling-PDF-Security-Fat"
         else
             failed_tests+=("Stirling-PDF-Security-Fat-Version-Check")
+            capture_failure_logs "Stirling-PDF-Security-Fat-Version-Check" "$CURRENT_CONTAINER"
             echo "Version verification failed for Stirling-PDF-Security-Fat"
         fi
+        stop_test_timer "Stirling-PDF-Security-Fat-Version-Check"
     fi
 
     docker-compose -f "./docker/embedded/compose/docker-compose-latest-fat-security.yml" down -v
@@ -617,10 +857,24 @@ main() {
         CUCUMBER_JUNIT_DIR="$PROJECT_ROOT/testing/cucumber/junit"
         mkdir -p "$CUCUMBER_JUNIT_DIR"
         cd "testing/cucumber"
+        start_test_timer "Stirling-PDF-Regression"
+
+        # Snapshot docker log line count before behave so we can extract only behave-window logs
+        DOCKER_LOG_BEFORE=$(docker logs "$CONTAINER_NAME" 2>&1 | wc -l)
+
+        export TEST_CONTAINER_NAME="$CONTAINER_NAME"
+        export TEST_REPORT_DIR="$REPORT_DIR"
+
+        gha_group "Test: Behave regression tests"
         if python -m behave \
             -f behave_html_formatter:HTMLFormatter -o "$CUCUMBER_REPORT" \
             -f pretty \
             --junit --junit-directory "$CUCUMBER_JUNIT_DIR"; then
+            gha_endgroup
+
+            # Save docker logs produced during the behave run
+            docker logs "$CONTAINER_NAME" 2>&1 | tail -n +"$((DOCKER_LOG_BEFORE + 1))" > "$REPORT_DIR/cucumber-docker-context.log" 2>/dev/null || true
+
             echo "Waiting 5 seconds for any file operations to complete..."
             sleep 5
 
@@ -636,10 +890,17 @@ main() {
             fi
             passed_tests+=("Stirling-PDF-Regression $CONTAINER_NAME")
         else
+            gha_endgroup
             failed_tests+=("Stirling-PDF-Regression $CONTAINER_NAME")
-            echo "Printing docker logs of failed regression"
+
-            docker logs "$CONTAINER_NAME"
+            # Save docker logs from the behave window to a dedicated file
-            echo "Printed docker logs of failed regression"
+            local cucumber_log="$REPORT_DIR/cucumber-docker-context.log"
+            docker logs "$CONTAINER_NAME" 2>&1 | tail -n +"$((DOCKER_LOG_BEFORE + 1))" > "$cucumber_log" 2>/dev/null || true
+            test_failure_logs["Stirling-PDF-Regression"]="$cucumber_log"
+
+            gha_group "Docker logs during behave run: $CONTAINER_NAME"
+            tail -100 "$cucumber_log"
+            gha_endgroup
 
             echo "Waiting 10 seconds before capturing file list..."
             sleep 10
@@ -648,6 +909,7 @@ main() {
             capture_file_list "$CONTAINER_NAME" "$AFTER_FILE"
             compare_file_lists "$BEFORE_FILE" "$AFTER_FILE" "$DIFF_FILE" "$CONTAINER_NAME"
         fi
+        stop_test_timer "Stirling-PDF-Regression"
     fi
     docker-compose -f "./docker/embedded/compose/test_cicd.yml" down -v
 
@@ -657,24 +919,32 @@ main() {
     run_tests "Stirling-PDF-Fat-Disable-Endpoints" "./docker/embedded/compose/docker-compose-latest-fat-endpoints-disabled.yml"
 
     if should_run_test "Disabled-Endpoints"; then
+        start_test_timer "Disabled-Endpoints"
+        gha_group "Test: Disabled-Endpoints"
         echo "Testing disabled endpoints..."
         if ./testing/test_disabledEndpoints.sh -f ./testing/endpoints.txt -b http://localhost:8080; then
             passed_tests+=("Disabled-Endpoints")
         else
             failed_tests+=("Disabled-Endpoints")
+            capture_failure_logs "Disabled-Endpoints" "$CURRENT_CONTAINER"
             echo "Disabled Endpoints tests failed"
         fi
+        gha_endgroup
+        stop_test_timer "Disabled-Endpoints"
     fi
 
     if should_run_test "Stirling-PDF-Fat-Disable-Endpoints-Version-Check"; then
+        start_test_timer "Stirling-PDF-Fat-Disable-Endpoints-Version-Check"
         echo "Testing version verification..."
         if verify_app_version "Stirling-PDF-Fat-Disable-Endpoints" "http://localhost:8080"; then
             passed_tests+=("Stirling-PDF-Fat-Disable-Endpoints-Version-Check")
             echo "Version verification passed for Stirling-PDF-Fat-Disable-Endpoints"
         else
             failed_tests+=("Stirling-PDF-Fat-Disable-Endpoints-Version-Check")
+            capture_failure_logs "Stirling-PDF-Fat-Disable-Endpoints-Version-Check" "$CURRENT_CONTAINER"
             echo "Version verification failed for Stirling-PDF-Fat-Disable-Endpoints"
         fi
+        stop_test_timer "Stirling-PDF-Fat-Disable-Endpoints-Version-Check"
     fi
 
     docker-compose -f "./docker/embedded/compose/docker-compose-latest-fat-endpoints-disabled.yml" down -v
@@ -682,24 +952,33 @@ main() {
     # ==================================================================
     # Final Report
     # ==================================================================
-    echo "All tests completed in $SECONDS seconds."
+    echo ""
+    echo "=========================================="
+    echo "TEST RESULTS SUMMARY"
+    echo "=========================================="
+    echo "Total duration: ${SECONDS}s"
+    echo "Passed: ${#passed_tests[@]}  Failed: ${#failed_tests[@]}"
+    echo ""
 
     if [ ${#passed_tests[@]} -ne 0 ]; then
         echo "Passed tests:"
         for test in "${passed_tests[@]}"; do
-            echo -e "\e[32m$test\e[0m"
+            local dur=${test_durations["$test"]:-"?"}
+            echo -e "  \e[32m✅ $test\e[0m (${dur}s)"
         done
     fi
 
     if [ ${#failed_tests[@]} -ne 0 ]; then
+        echo ""
         echo "Failed tests:"
         for test in "${failed_tests[@]}"; do
-            echo -e "\e[31m$test\e[0m"
+            local dur=${test_durations["$test"]:-"?"}
+            local log=${test_failure_logs["$test"]:-"no log captured"}
+            echo -e "  \e[31m❌ $test\e[0m (${dur}s) -> $log"
         done
     fi
 
-    # Save failed tests for potential rerun
+    echo ""
-    save_failed_tests
 
     if [ ${#failed_tests[@]} -ne 0 ]; then
         echo "Some tests failed."