dragonfly accesses the docker image on Google？

[zsksy123]

version information

NAME     	NAMESPACE	REVISION	UPDATED                             	STATUS  	CHART           	APP VERSION
dragonfly	dragonfly	17      	2025-04-22 11:10:55.795276 +0800 CST	deployed	dragonfly-1.1.32	2.1.31

My configuration is as follows

dfdaemon:
    config:
        resourceClients:
            https:
                proxy: "http://10.96.2.59:7890"
            http:
                proxy: "http://10.96.2.59:7890"

However, it has not taken effect. Is there any problem with the configuration? I hope you can correct it.

@matiasinsaurralde @bergwolf @liubin @jim3ma Have you guys continued to maintain this project?

[gaius-qi]

You can use the new version charts, refer to https://d7y.io/docs/next/getting-started/quick-start/kubernetes/#create-dragonfly-cluster-based-on-helm-charts. @zsksy123

[zsksy123]

@gaius-qi Thank you for your reply. What I mean is that for some foreign mirror sites, can Dragonfly use the proxy I configured to pull content? Or, Drafongly automatically injects this proxy into certain configurations. When pulling images, will the overseas servers use this proxy?

Additionally, it seems that my configuration has not taken effect. The configuration is as follows.

client:
  enable: true
    config:
      # -- verbose prints log.
      verbose: true
      log:
        # -- Specify the logging level [trace, debug, info, warn, error]
        level: info
      proxy:
        # -- addr is the proxy server address of dfdaemon.
        addr: http://127.0.0.1:4001
      containerRuntime:
        containerd:
          # -- configPath is the path of containerd configuration file.
          configPath: /etc/containerd/config.toml
          registries:
            - hostNamespace: docker.io
              serverAddr: https://index.docker.io
              capabilities: ["pull", "resolve"]
              skipVerify: true
              # ca: []
            - hostNamespace: ghcr.io
              serverAddr: https://ghcr.io
              capabilities: ["pull", "resolve"]
              skipVerify: true
              # ca: []
            - hostNamespace: gcr.io
              serverAddr: https://gcr.io
              capabilities: ["pull", "resolve"]
              skipVerify: true
              # ca: []

The version information is as follows

[zsksy123]

@gaius-qi Let me briefly describe my requirement. Could I inject an env variable into dragonfly to pull the image from Google? Is there such a configuration?

[LunaWhispers]

@zsksy123 The first time the client downloads a file, the scheduler will let the seed peer download from the source back to the source, following the normal http download logic. You don't need to configure extra configuration for different sources. If you want to download the file from Google, the source station here can be Google. If you have configured the dragonfly environment correctly, please check if the seed peer node can ping through Google.