From 243ab7d720cfb3bc25b1e71689b0834b8f5afd01 Mon Sep 17 00:00:00 2001 From: Li Wang Date: Fri, 19 Dec 2025 21:21:42 +0800 Subject: [PATCH] [CI] Use offline mode for nightly test (#5187) ### What this PR does / why we need it? For single node test, the lack of a retry mechanism for accessing ModelScope resulted in an HTTP 400 error sometimes. I recommend using a local offline cache instead. - vLLM version: v0.12.0 - vLLM main: https://github.com/vllm-project/vllm/commit/ad32e3e19ccf0526cb6744a5fed09a138a5fb2f9 --------- Signed-off-by: wangli --- .github/workflows/_e2e_nightly_single_node.yaml | 3 +++ .github/workflows/pr_tag_image_build_and_push.yaml | 2 +- tests/e2e/conftest.py | 2 +- 3 files changed, 5 insertions(+), 2 deletions(-) diff --git a/.github/workflows/_e2e_nightly_single_node.yaml b/.github/workflows/_e2e_nightly_single_node.yaml index df91260f..bff4f0e2 100644 --- a/.github/workflows/_e2e_nightly_single_node.yaml +++ b/.github/workflows/_e2e_nightly_single_node.yaml @@ -57,6 +57,9 @@ jobs: timeout-minutes: 600 container: image: ${{ inputs.image }} + env: + TRANSFORMERS_OFFLINE: 1 + VLLM_USE_MODELSCOPE: True steps: - name: Check npu and CANN info run: | diff --git a/.github/workflows/pr_tag_image_build_and_push.yaml b/.github/workflows/pr_tag_image_build_and_push.yaml index 4991ae04..a417fb2c 100644 --- a/.github/workflows/pr_tag_image_build_and_push.yaml +++ b/.github/workflows/pr_tag_image_build_and_push.yaml @@ -26,7 +26,7 @@ on: - 'cmake/**' - 'CMakeLists.txt' - 'csrc/**' - types: [ labeled, synchronize ] + types: [ labeled ] push: # Publish image when tagging, the Dockerfile in tag will be build as tag image branches: diff --git a/tests/e2e/conftest.py b/tests/e2e/conftest.py index a8207e3f..b59ab7ce 100644 --- a/tests/e2e/conftest.py +++ b/tests/e2e/conftest.py @@ -162,7 +162,7 @@ class RemoteOpenAIServer: self.proxy_port = proxy_port self._start_server(model, vllm_serve_args, env_dict) - max_wait_seconds = max_wait_seconds or 1800 + max_wait_seconds = max_wait_seconds or 2800 if self.disaggregated_prefill: assert proxy_port is not None, "for disaggregated_prefill, proxy port must be provided" self._wait_for_server_pd(timeout=max_wait_seconds)