gahow/aituner
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Match Qwen30 probe file descriptor limit

Browse Source
This commit is contained in:
Gahow Wang
2026-07-19 01:07:57 +08:00
parent 6726318792
commit 9837fa5133
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
runs/frontier-fidelity-envelope-v1/run_qwen30_fixed_pd_pressure_probe.sh
View File

@@ -31,6 +31,12 @@ SERVER_PID=""
mkdir -p "${OUT}/provenance" "${OUT}/trials" "${FLASHINFER_WORKSPACE_BASE}" mkdir -p "${OUT}/provenance" "${OUT}/trials" "${FLASHINFER_WORKSPACE_BASE}"
exec > >(tee -a "${OUT}/controller.log") 2>&1 exec > >(tee -a "${OUT}/controller.log") 2>&1
# Match the validated Qwen30 real-surface runner's file-descriptor budget for
# vLLM's multiprocessing shared-memory transport. The failed r1 probe had the
# default remote limit (1024) and stalled before KV-cache creation; r2 tests
# whether this runner-contract difference is causal.
ulimit -n 65536
cleanup_server() { cleanup_server() {
if [[ -n "${SERVER_PID}" ]] && kill -0 "${SERVER_PID}" 2>/dev/null; then if [[ -n "${SERVER_PID}" ]] && kill -0 "${SERVER_PID}" 2>/dev/null; then
kill -TERM -- "-${SERVER_PID}" 2>/dev/null || true kill -TERM -- "-${SERVER_PID}" 2>/dev/null || true
@@ -204,6 +210,7 @@ PY
sha256sum "${BASH_SOURCE[0]}" "${CLIENT}" "${MODEL_ROOT}/config.json" > "${OUT}/provenance/input.sha256" sha256sum "${BASH_SOURCE[0]}" "${CLIENT}" "${MODEL_ROOT}/config.json" > "${OUT}/provenance/input.sha256"
"${VENV_ROOT}/bin/vllm" --version > "${OUT}/provenance/vllm.version" "${VENV_ROOT}/bin/vllm" --version > "${OUT}/provenance/vllm.version"
"${VENV_ROOT}/bin/python" -c 'import torch, transformers, vllm; print(f"torch={torch.__version__}"); print(f"transformers={transformers.__version__}"); print(f"vllm={vllm.__version__}")' > "${OUT}/provenance/runtime.versions" "${VENV_ROOT}/bin/python" -c 'import torch, transformers, vllm; print(f"torch={torch.__version__}"); print(f"transformers={transformers.__version__}"); print(f"vllm={vllm.__version__}")' > "${OUT}/provenance/runtime.versions"
ulimit -n > "${OUT}/provenance/open-file-limit"
nvidia-smi --query-gpu=index,name,uuid,driver_version,memory.total --format=csv,noheader > "${OUT}/provenance/gpus.before.csv" nvidia-smi --query-gpu=index,name,uuid,driver_version,memory.total --format=csv,noheader > "${OUT}/provenance/gpus.before.csv"
declare -a ORDERS=("4 8 12 16" "16 12 8 4" "8 16 4 12") declare -a ORDERS=("4 8 12 16" "16 12 8 4" "8 16 4 12")