飞桨云端V100机子该如何进行分布式训练，不知道该如何使得两机子相互ping通

[WuHui6]

请提供下述完整信息以便快速定位问题

• 系统环境：飞桨云端BML Codelab

• 版本号　Paddle：2.2.2 PaddleOCR：

• 运行指令：python3 -m paddle.distributed.launch --log_dir=./log/ --ips="10.32.116.210,10.32.167.53" --gpus="0,1" tools/train.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml

• 完整报错：
  INFO 2024-03-08 14:53:46,608 launch_utils.py:532] details abouts PADDLE_TRAINER_ENDPOINTS can be found in ./log//endpoints.log, and detail running logs maybe found in ./log//workerlog.0
  launch proc_id:2480 idx:0
  launch proc_id:2485 idx:1
  I0308 14:53:49.568403 2480 gen_comm_id_helper.cc:190] Server listening on: 10.32.167.53:6070 successful.
  INFO 2024-03-08 14:53:52,677 launch_utils.py:320] terminate process group gid:2480
  INFO 2024-03-08 14:53:56,680 launch_utils.py:341] terminate all the procs
  ERROR 2024-03-08 14:53:56,680 launch_utils.py:604] ABORT!!! Out of all 4 trainers, the trainer process with rank=[3] was aborted. Please check its log.
  INFO 2024-03-08 14:54:00,681 launch_utils.py:341] terminate all the procs
  INFO 2024-03-08 14:54:00,682 launch.py:311] Local processes completed.

[WuHui6]

飞桨云端机子可以进行分布式训练吗？