# Only head node started, not worker nodes

**URL:** <https://discuss.ray.io/t/only-head-node-started-not-worker-nodes/4532>\
**Category:** Ray Clusters\
**Created:** [December 28, 2021, 9:48pm UTC](https://discuss.ray.io/t/only-head-node-started-not-worker-nodes/4532 "2021-12-28T21:48:01Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![hokyjack](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/hokyjack/32/1922_2.png) [@hokyjack](https://discuss.ray.io/u/hokyjack)\
**Post date:** [December 28, 2021, 9:48pm UTC](https://discuss.ray.io/t/only-head-node-started-not-worker-nodes/4532/1 "2021-12-28T21:48:01Z")

</div>

Hello, I have a problem with starting an autoscaling cluster on AWS.  
I start the cluster with my YAML file, but then, no worker nodes are started. What I am doing wrong please?

Ray version 1.9.1 on WSL.  
Python 3.7

```auto
# A unique identifier for the head node and workers of this cluster.
cluster_name: basic-ray3
# The maximum number of workers nodes to launch in addition to the head
# node. This takes precedence over min_workers. min_workers defaults to 0.

upscaling_speed: 5.0

max_workers: 12

#idle_timeout_minutes: 5

available_node_types:
  ray.head.default:
    resources: {"CPU": 4}
    node_config:
      InstanceType: m5.xlarge
      KeyName: hoky-ray
      #ImageId: latest_dlami

  ray.worker.default:
    resources: {"CPU": 4}
    min_workers: 12
    max_workers: 12
    node_config:
      InstanceType: m5.xlarge
      KeyName: hoky-ray
      #ImageId: latest_dlami

# Cloud-provider specific configuration.
provider:
   type: aws
   region: us-west-2
   availability_zone: us-west-2a

# How Ray will authenticate with newly launched nodes.
auth:
   ssh_user: ubuntu
   ssh_private_key: ~/.ssh/hoky-ray.pem

setup_commands:
  - pip install ray[all] # We won’t use pytorch. 
# However, this and the following line demonstrate that you can specify arbitrary
# startup scripts on the cluster.
  - pip install empyrical pandas==1.2.3 tqdm

file_mounts: {
   "~": ".", # /mnt/c/Users/Hoky/ray
}

head_node_type: ray.head.default
worker_default_node_type: ray.worker.default

# Command to start ray on the head node. You don't need to change this.
head_start_ray_commands:
    - ray stop
    - ulimit -n 65536; ray start --head --port=6379 --object-manager-port=8076 --autoscaling-config=~/ray_bootstrap_config.yaml

# Command to start ray on worker nodes. You don't need to change this.
worker_start_ray_commands:
    - ray stop
    - ulimit -n 65536; ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076

```

 ![image](https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/e/e4a74ce86966e56b22debbdc351694469dde9d90.png)

```auto
hoky@DESKTOP-P1L6T2J:/mnt/c/Users/Hoky/ray$ ray up cluster.yaml -y && ray submit cluster.yaml compute_remote.py
Cluster: basic-ray3

2021-12-28 22:28:40,196 INFO util.py:282 -- setting max workers for head node type to 0
Checking AWS environment settings
AWS config
  IAM Profile: ray-autoscaler-v1 [default]
  EC2 Key pair (all available node types): hoky-ray
  VPC Subnets (all available node types): subnet-071195b1ea481cc82 [default]
  EC2 Security groups (all available node types): sg-030fd1153af5ebf9d [default]
  EC2 AMI (all available node types): ami-0a2363a9cff180a64 [dlami]

No head node found. Launching a new cluster. Confirm [y/N]: y [automatic, due to --yes]

Acquiring an up-to-date head node
  Launched 1 nodes [subnet_id=subnet-071195b1ea481cc82]
    Launched instance i-00186169f828c3616 [state=pending, info=pending]
  Launched a new head node
  Fetching the new head node

<1/1> Setting up head node
  Prepared bootstrap config
  New status: waiting-for-ssh
  [1/7] Waiting for SSH to become available
    Running `uptime` as a test.
    Waiting for IP
      Not yet available, retrying in 5 seconds
      Received: 52.40.45.139

```

Thanks for the tips!

---

<div class="post-metadata">

**Author:** ![ckw017](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/ckw017/32/1683_2.png) [@ckw017](https://discuss.ray.io/u/ckw017)\
**Post date:** [January 19, 2022, 1:04am UTC](https://discuss.ray.io/t/only-head-node-started-not-worker-nodes/4532/2 "2022-01-19T01:04:10Z")

</div>

A different user had a similar issue [here](https://github.com/ray-project/ray/issues/19633) with the cluster feeling to create worker nodes, the solution was to update service account roles. Can you check if you’re seeing similar output from the scheduler