FederatedAI / KubeFATE

Manage federated learning workload using cloud native technologies.
Apache License 2.0
424 stars 221 forks source link

KubeFATE v1.7.0 docker-compose-eggroll.yml 中的挂载文件没有配置到真正的路径 #940

Open AmazingTai opened 6 months ago

AmazingTai commented 6 months ago

What deployment mode you are use

  1. docker-compose;

What KubeFATE and FATE version you are using v1.7.0

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

What happen?

  python:
    image: "federatedai/python:${TAG}"
    environment:
      PYTHONPATH: "$PYTHONPATH:/data/projects/fate/fate/python:/data/projects/fate/eggroll/python:/data/projects/fate/fateflow/python:/data/projects/fate/fate/python/fate_client"
      FATE_PROJECT_BASE: "/data/projects/fate"
    ports:
      - "9360:9360"
      - "9380:9380"
    restart: always
    volumes:
      - shared_dir_federatedml:/data/projects/fate/fate/python/federatedml
      - shared_dir_examples:/data/projects/fate/examples
      - download_dir:/data/projects/fate/python/download_dir
      - ./confs/fate_flow/conf:/data/projects/fate/conf
      ......

重点是- ./confs/fate_flow/conf:/data/projects/fate/conf 当我修改./confs/fate_flow/conf/job_default_config.yaml时,我发现提交的任务并没有使用job_default_config中的配置。然后我进入python容器,进入/data/projects/fate/,发现/fateflow/conf还有一个job_default_config.yaml。修改此处配置重启容器后配置生效。 经过修python/fate_flow/db/job_default_config.py改代码打点:

image

在std.log中打印发型使用的是/data/projects/fate/fateflow/conf/job_default_config配置。 image 也就是说docker-compose-eggroll.yml中的- ./confs/fate_flow/conf:/data/projects/fate/conf 并没有正确挂载。应该是:

To Validate 为了验证猜想,我发现在v1.9: