.env.example

# Ray Cluster Configuration
# Copy this file to .env and update with your cluster details

# Required: Ray cluster address
# Format: ray://head-node-ip:port or http://head-node-ip:8265
# Example: ray://192.168.1.100:10001
RAY_ADDRESS=ray://your-cluster-head:10001

# Required: Namespace for logical grouping of jobs and actors
# Example: production, development, staging
RAY_NAMESPACE=production

# Optional: Runtime environment as JSON string
# Example: {\"pip\": [\"numpy\", \"pandas\"], \"env_vars\": {\"MY_VAR\": \"value\"}}
RAY_RUNTIME_ENV=

# Optional: Job configuration as JSON string
# Example: {\"runtime_env\": {\"pip\": [\"requests\"]}}
RAY_JOB_CONFIG=

# Optional: Maximum task retries (default: 3)
# Set to 0 to disable retries
RAY_TASK_MAX_RETRIES=3

# Optional: GCS reconnection timeout in seconds
RAY_gcs_rpc_server_reconnect_timeout_s=60

# Optional: Enable pickle debugging (0=off, 1=basic, 2=verbose)
RAY_PICKLE_VERBOSE_DEBUG=0