From bccd2d0c959e439ac4f2dcc5a1b5f5a21ea7dc08 Mon Sep 17 00:00:00 2001 From: samzong Date: Tue, 14 Oct 2025 16:45:44 +0800 Subject: [PATCH] refactor(e2e): remove legacy mock/real vLLM test modes and Makefile targets Signed-off-by: samzong --- e2e-tests/run_all_tests.py | 18 ---------------- tools/make/build-run-test.mk | 42 +----------------------------------- 2 files changed, 1 insertion(+), 59 deletions(-) diff --git a/e2e-tests/run_all_tests.py b/e2e-tests/run_all_tests.py index 16fdf686..b3147be7 100644 --- a/e2e-tests/run_all_tests.py +++ b/e2e-tests/run_all_tests.py @@ -89,12 +89,6 @@ def main(): ) parser.add_argument("--pattern", default="*.py", help="Test file pattern to run") parser.add_argument("--verbose", "-v", action="store_true", help="Verbose output") - parser.add_argument( - "--mock", action="store_true", help="Running with mock vLLM servers" - ) - parser.add_argument( - "--real", action="store_true", help="Running with real vLLM servers" - ) args = parser.parse_args() # Get the directory where this script is located @@ -127,18 +121,6 @@ def main(): print(f"No test files found matching pattern '{args.pattern}'") return 1 - # Print test mode information - if args.mock: - print("\n🤖 Running in MOCK mode - using mock vLLM servers") - print(" ✅ Fast execution, no GPU required") - print(" ⚠️ Mock responses, not real model inference") - elif args.real: - print("\n🧠 Running in REAL mode - using actual vLLM servers") - print(" 🚀 Real model inference and responses") - print(" ⚠️ Requires GPU and longer execution time") - else: - print("\n🔍 Running in STANDARD mode - checking whatever is available") - print(f"\nRunning {len(test_files)} test files:") for file in test_files: print(f" - {file}") diff --git a/tools/make/build-run-test.mk b/tools/make/build-run-test.mk index 64fc68db..eba135e7 100644 --- a/tools/make/build-run-test.mk +++ b/tools/make/build-run-test.mk @@ -113,50 +113,10 @@ start-llm-katan: @echo "Press Ctrl+C to stop servers" @./e2e-tests/start-llm-katan.sh -# Legacy: Start mock vLLM servers for testing (foreground mode for development) -start-mock-vllm: - @echo "Starting mock vLLM servers in foreground mode..." - @echo "Press Ctrl+C to stop servers" - @./e2e-tests/start-mock-servers.sh - -# Start real vLLM servers for testing -start-vllm: - @echo "Starting real vLLM servers..." - @./e2e-tests/start-vllm-servers.sh - -# Stop real vLLM servers -stop-vllm: - @echo "Stopping real vLLM servers..." - @./e2e-tests/stop-vllm-servers.sh - # Run e2e tests with LLM Katan (lightweight real models) test-e2e-vllm: @echo "Running e2e tests with LLM Katan servers..." @echo "⚠️ Note: Make sure LLM Katan servers are running with 'make start-llm-katan'" @python3 e2e-tests/run_all_tests.py -# Legacy: Run e2e tests with mock vLLM (assumes mock servers already running) -test-e2e-mock: - @echo "Running e2e tests with mock vLLM servers..." - @echo "⚠️ Note: Make sure mock servers are running with 'make start-mock-vllm'" - @python3 e2e-tests/run_all_tests.py --mock - -# Run e2e tests with real vLLM (assumes real servers already running) -test-e2e-real: - @echo "Running e2e tests with real vLLM servers..." - @echo "⚠️ Note: Make sure real vLLM servers are running with 'make start-vllm'" - @python3 e2e-tests/run_all_tests.py --real - - -# Note: Automated tests not supported with foreground-only mock servers -# Use the manual workflow: make start-llm-katan in one terminal, then run tests in another - -# Full automated test with cleanup (for CI/CD) -test-e2e-real-automated: start-vllm - @echo "Running automated e2e tests with real vLLM servers..." - @sleep 5 - @python3 e2e-tests/run_all_tests.py --real || ($(MAKE) stop-vllm && exit 1) - @$(MAKE) stop-vllm - -# Run all e2e tests (LLM Katan, mock and real) -test-e2e-all: test-e2e-vllm test-e2e-mock test-e2e-real +# Note: Use the manual workflow: make start-llm-katan in one terminal, then run tests in another