alibaba / alibaba/ROLL

Failed to register worker to Raylet: IOError

Open
#263 6 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
3.4k
Forks
312
Avg merge
1h 2m
Merged PRs (30d)
2

Description

```
[2025-11-23 22:52:41] [initialize.py (46)] [INFO] [DRIVER 0 / 1][PID 51567] Starting ray cluster: ray start --head --port=6379 --node-name=127.0.0.1:0 --dashboard-port=8265
2025-11-23 22:52:47,374 INFO worker.py:1837 -- Connecting to existing Ray cluster at address: [2605:340:cd51:6900:3829:8af0:4e1f:1ffb]:6379...
2025-11-23 22:52:47,385 INFO worker.py:2023 -- Connected to Ray cluster.
[2025-11-23 22:52:54,526 E 51567 51567] core_worker_process.cc:223: Failed to register worker to Raylet: IOError: Failed to read data from the socket: End of file worker_id=01000000ffffffffffffffffffffffffffffffffffffffffffffffff
```
请问这个问题该如何处理呢?
我的启动脚本如下
```
sudo pip install -r requirements_common.txt
export PYTHONPATH=$(pwd):$PYTHONPATH

python examples/start_rlvr_pipeline.py \
--config_path qwen3-8B-rlvr_megatron \
--config_name rlvr_config
```

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.