kubeFate example upload data to hdfs failed

[jerryhong0307]

**What deployment mode you are use? ** Kuberentes --> minikube version: v1.20.0

**What KubeFATE and FATE version you are using? ** KubeFATE: v 1.4.5 FATE: v 1.9.0 MUST Please state the KubeFATE and FATE version you found the issue

**What OS you are using for docker-compse or Kubernetes? Please also clear the version of OS. **

• OS: Ubuntu
• Version : 22.04

参照流程: https://github.com/FederatedAI/KubeFATE/blob/master/docs/tutorials/Build_Two_Parties_FATE_Cluster_in_One_Linux_Machine_with_MiniKube.md

并于 fate-9999 pipeline_tutorial_upload_jerry.ipynb 中执行到pipeline_upload.upload(drop=1) 资料的时候会有 hdfs connection 相关问题, fateboard job log:

手动于 k8s_fateflow_python-0_fate-9999_a661fab6-509a-4893-b6c8-cc3f6a6a94b5_0 container 中加入 pyarrow & hdfs client 与环境变数皆无法正常执行 upload 动作.

以下为 k8s_fateflow_python-0_fate-9999_a661fab6-509a-4893-b6c8-cc3f6a6a94b5_0 中修改的环境变数:

• ~/.bashrc

• /data/projects/fate/conf/service_conf.yaml

• $HADOOP_HOME/etc/hadoop/core-site.xml

请问一下应该怎么解决呢？谢谢

[JingChen23]

Please get into your hadoop namenode pod, and run hdfs dfsadmin -report, and check if the ip address of your datanode is the pod ip of your datanode