Skip to content

[codex] Truncate inference rank logs with tee #2817

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
samsja wants to merge 1 commit into
base: main
Choose a base branch
from
+6 −4
Draft

[codex] Truncate inference rank logs with tee #2817

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions src/prime_rl/templates/_launch_rank.sh.j2
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,14 +15,18 @@ rank_log() {
# launch_inference_rank <port> <cuda_visible_devices> <dp> <rpc_port> <vllm_extra_json|""> <nixl_side_channel_port|""> <log_file>
launch_inference_rank() {
local port="$1" gpus="$2" dp="$3" rpc="$4" extra="$5" nixl="$6" log="$7"
local log_name="${log##*/}"
local -a tee_args=(-a "$log")
# node_*.log already gets a truncating header; rank-specific logs do not.
[[ "$log_name" == *_rank*.log ]] && tee_args=("$log")
local -a cmd=(uv run inference @ "$CONFIG_PATH"
--server.host 0.0.0.0 --server.port "$port"
--parallel.dp "$dp" --data-parallel-size-local 1 --api-server-count 1)
[ -n "$rpc" ] && cmd+=(--data-parallel-rpc-port "$rpc")
[ -n "$extra" ] && cmd+=(--vllm-extra "$extra")
if [ -n "$nixl" ]; then
CUDA_VISIBLE_DEVICES="$gpus" VLLM_NIXL_SIDE_CHANNEL_PORT="$nixl" "${cmd[@]}" 2>&1 | tee -a "$log" &
CUDA_VISIBLE_DEVICES="$gpus" VLLM_NIXL_SIDE_CHANNEL_PORT="$nixl" "${cmd[@]}" 2>&1 | tee "${tee_args[@]}" &
else
CUDA_VISIBLE_DEVICES="$gpus" "${cmd[@]}" 2>&1 | tee -a "$log" &
CUDA_VISIBLE_DEVICES="$gpus" "${cmd[@]}" 2>&1 | tee "${tee_args[@]}" &
fi
}
1 change: 0 additions & 1 deletion src/prime_rl/templates/inference.sbatch.j2
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,6 @@ export PROJECT_DIR={{ project_dir }}
export CONFIG_PATH={{ config_path }}
export OUTPUT_DIR={{ output_dir }}
mkdir -p $OUTPUT_DIR/logs/inference
rm -f $OUTPUT_DIR/logs/inference/*.log
ln -sfn inference/node_0.log $OUTPUT_DIR/logs/inference.log

# General
Expand Down
1 change: 0 additions & 1 deletion src/prime_rl/templates/multi_node_rl.sbatch.j2
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,7 +57,6 @@ export CONFIG_DIR={{ config_dir }}
export OUTPUT_DIR={{ output_dir }}
export ORCHESTRATOR_OUTPUT_DIR={{ orchestrator_output_dir }}
mkdir -p $OUTPUT_DIR/logs/trainer $OUTPUT_DIR/logs/inference
rm -f $OUTPUT_DIR/logs/inference/*.log
ln -sfn trainer/node_0.log $OUTPUT_DIR/logs/trainer.log
ln -sfn inference/node_0.log $OUTPUT_DIR/logs/inference.log

Expand Down
Loading