[Nightly] Optimize nightly CI (#4509)
### What this PR does / why we need it? 1. Optimize multi-node waiting logic 2. Remove the `tee` pipeline for logs, which will lead to hang issue ### How was this patch tested? - vLLM version: v0.12.0 - vLLM main: https://github.com/vllm-project/vllm/commit/v0.12.0 Signed-off-by: wangli <wangli858794774@gmail.com>
This commit is contained in:
@@ -78,6 +78,7 @@ if [ -n "$LOCAL_DEVICE_IDS" ]; then
|
||||
fi
|
||||
|
||||
if [[ -n "${GEN_RANKTABLE}" || ! -e ${PWD}/ranktable.json ]]; then
|
||||
timeout 180s \
|
||||
GLOO_SOCKET_IFNAME=$NETWORK_CARD_NAME torchrun \
|
||||
--nproc_per_node 1 \
|
||||
--nnodes ${NNODES} \
|
||||
|
||||
Reference in New Issue
Block a user