#40566·ray

[Clusters] rsync incompatibility error when using other docker images

Author: fehtemamCreated Oct 23, 2023Updated Sep 17, 2026

LabelsbugenhancementP2corecore-clusters

What happened + What you expected to happen

When building a cluster from the culster.yaml file, if we use an image other than the ray docker image, rsync throws an error of:

protocol version mismatch -- is your shell clean? (see the rsync man page for an explanation) rsync error: protocol incompatibility (code 2) at compat.c(178) [sender=3.1.2]

I have not seen in the documentation anything about the ray docker image being the only option to use with clusters. The expected behavior is to have the cluster started with no errors.

Versions / Dependencies

Python: 3.10.10 Ray: 2.6.3

Reproduction script

Here is an example yaml file to reproduce:

yaml

cluster_name: d_frcst
max_workers: 6
upscaling_speed: 1.0
docker:
    image: ubuntu:22.04
    # image: rayproject/ray:latest-cpu 
    container_name: "ray_container"
    pull_before_run: True
    run_options: 
        - --ulimit nofile=65536:65536
idle_timeout_minutes: 5
provider:
    type: aws
    region: us-east-1
    cache_stopped_nodes: False

auth:
    ssh_user: ubuntu

available_node_types:
    head:
        node_config:
            InstanceType: c6a.large
            BlockDeviceMappings:
                - DeviceName: /dev/sda1
                  Ebs:
                      VolumeSize: 140
    worker:
        min_workers: 5
        max_workers: 6
        node_config:
            InstanceType: c6a.16xlarge
            InstanceMarketOptions:
                MarketType: spot
head_node_type: head

head_start_ray_commands:
    - ray stop
    - ray start --head --port=6379 --object-manager-port=8076
worker_start_ray_commands:
    - ray stop
    - ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076


### Issue Severity

High: It blocks me from completing my task.

Source: ray-project/ray

View original on GitHub View discussion on GitHub