ワンクリックでManusで任意のスキルを実行

ray-start

スター4

フォーク1

更新日2026年5月23日 03:10

Start the Ray cluster across all 4 DGX Spark nodes. Copies the per-node start-ray scripts, starts spark-01 as head, then joins spark-02/03/04 with 10-second pauses between each. Verifies cluster health with ray status afterward.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

urbanspr1nter

urbanspr1nter/dgx-spark-bare-metal

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

Ray Cluster Start

Start the Ray cluster on all 4 DGX Spark nodes in the correct order with proper pacing.

What It Does

Copy scripts — Copies each cluster/start-ray-spark-*.sh to $HOME/models/start-ray.sh on the corresponding node.
Start spark-01 (head) — Runs start-ray.sh from the vllm directory (where .venv lives) on spark-01.
Wait 10 seconds.
Start spark-02 — Same pattern, via SSH.
Wait 10 seconds. Repeat for spark-03, then spark-04.
Verify — Runs ray status on spark-01 to confirm all nodes are active.

Usage

./scripts/start_cluster.sh

Run from the skill directory, or from anywhere — the script locates the repo root and env.sh automatically.

Stopping the Cluster

CRITICAL — Stop in reverse order: spark-04 → spark-03 → spark-02 → spark-01. Workers must be stopped before the head node.

./scripts/stop_cluster.sh

Prerequisites

vLLM installed at $HOME/models/vllm/.venv on all nodes.
Passwordless SSH to all Sparks.
env.sh at repo root with SPARK_01_IP through SPARK_04_IP defined.
cluster/start-ray-spark-{a,b,c,d}.sh scripts present in the repo.

Before Reinstalling vLLM

CRITICAL — Always stop the Ray cluster before running clean_vllm.sh or reinstalling vLLM. Wiping the venv while Ray is running will leave the cluster in a broken state.

このリポジトリの他の Skills

同じリポジトリ

vllm-install

urbanspr1nter/dgx-spark-bare-metal

Install or reinstall vLLM across the entire 4× DGX Spark cluster. Builds vLLM from source on all nodes, using the setup/install_vllm.sh script. Handles rsync-ing the repo, kicking off builds via tmux, and verifying the binary exists on every node afterward.

2026-05-244

cluster-diagnostics

urbanspr1nter/dgx-spark-bare-metal

Run diagnostics on the 4x DGX Spark cluster. Checks node reachability via SSH, verifies model directories exist on all nodes, and validates that paths referenced in model run scripts actually exist on each node. Use before starting Ray or running models, or when troubleshooting cluster issues.

2026-05-234

name	ray-start
description	Start the Ray cluster across all 4 DGX Spark nodes. Copies the per-node start-ray scripts, starts spark-01 as head, then joins spark-02/03/04 with 10-second pauses between each. Verifies cluster health with ray status afterward.

Ray Cluster Start

Start the Ray cluster on all 4 DGX Spark nodes in the correct order with proper pacing.

What It Does

Copy scripts — Copies each cluster/start-ray-spark-*.sh to $HOME/models/start-ray.sh on the corresponding node.
Start spark-01 (head) — Runs start-ray.sh from the vllm directory (where .venv lives) on spark-01.
Wait 10 seconds.
Start spark-02 — Same pattern, via SSH.
Wait 10 seconds. Repeat for spark-03, then spark-04.
Verify — Runs ray status on spark-01 to confirm all nodes are active.

Usage

./scripts/start_cluster.sh

Run from the skill directory, or from anywhere — the script locates the repo root and env.sh automatically.

Stopping the Cluster

CRITICAL — Stop in reverse order: spark-04 → spark-03 → spark-02 → spark-01. Workers must be stopped before the head node.

./scripts/stop_cluster.sh

Prerequisites

vLLM installed at $HOME/models/vllm/.venv on all nodes.
Passwordless SSH to all Sparks.
env.sh at repo root with SPARK_01_IP through SPARK_04_IP defined.
cluster/start-ray-spark-{a,b,c,d}.sh scripts present in the repo.

Before Reinstalling vLLM

CRITICAL — Always stop the Ray cluster before running clean_vllm.sh or reinstalling vLLM. Wiping the venv while Ray is running will leave the cluster in a broken state.