fix

.github/workflows/pylint.yml

@@ -29,4 +29,13 @@ jobs:
           -v ${PWD}:/workspace \
           -w /workspace \
           registry.cn-beijing.aliyuncs.com/llumnix/llumnix-dev:20240909_action_678a439 \
-          bash -c "pip install -e . > /dev/null && make lint"
+          bash -c "pip install -e . > /dev/null && make lint"
+    - uses: actions/github-script@v7
+      with:
+        script: |
+          github.rest.issues.createComment({
+            issue_number: context.issue.number,
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            body: '**ok**'
+          })

tests/e2e_test/test_bench.py

@@ -124,4 +124,4 @@ async def run_bench_command(command):
 
     shutdown_llumnix_service()
     clear_ray_state()
-    await asyncio.sleep(10)
+    await asyncio.sleep(3)

tests/e2e_test/test_e2e.py

@@ -106,18 +106,19 @@ async def test_e2e(model):
     }
 
     # generate llumnix outputs
+    shutdown_llumnix_service()
     launch_llumnix_service(model, max_model_len=max_model_len)
     await asyncio.sleep(60)
 
     llumnix_output = {}
     for prompt in prompts:
         response = await asyncio.wait_for(get_llumnix_responce(prompt, sampling_params, "127.0.0.1:37000"),
-                                          timeout=60*2)
+                                          timeout=60*5)
         llumnix_output[prompt] = response['text'][0]
 
     shutdown_llumnix_service()
     clear_ray_state()
-    await asyncio.sleep(10)
+    await asyncio.sleep(5)
 
     # get raw vllm outputs
     raw_vllm = LLM(model=model, trust_remote_code=True, max_model_len=max_model_len)

tests/e2e_test/test_migration.py

@@ -84,4 +84,4 @@ async def run_bench_command(command):
 
     shutdown_llumnix_service()
     clear_ray_state()
-    await asyncio.sleep(10)
+    await asyncio.sleep(3)