Admin quickly uses the container registry to serve the platform¶
In DCE 5.0, there are two methods for managing container registries: Managed Harbor (self-built Harbor on DCE) and Integrated Registry (integrating external Harbor or Docker Registry).
- Managed Harbor(suggested)
- Integrated Registry
Harbor is mainly promoted as the Container Registry in DCE 5.0 to provide image services.
Sharing public images¶
Assuming that you have created a Managed Harbor or integrated to an external harbor, follow the steps below to share a public image with all namespaces:
graph TB
create[Create managed Harbor] --> setpublic[Create a registry space and set it to public]
--> push[Push image]
--> deploy[Deploy application]
classDef plain fill:#ddd,stroke:#fff,stroke-width:1px,color:#000;
classDef k8s fill:#326ce5,stroke:#fff,stroke-width:1px,color:#fff;
classDef cluster fill:#fff,stroke:#bbb,stroke-width:1px,color:#326ce5;
class create,setpublic,push,deploy cluster;
click create "https://docs.daocloud.io/en/kangaroo/hosted/"
click setpublic "https://docs.daocloud.io/en/kangaroo/integrate/create-space/"
click push "https://docs.daocloud.io/en/kangaroo/quickstart/push/"
click deploy "https://docs.daocloud.io/en/kpanda/user-guide/workloads/create-deployment/"
Expected Result: When users on the platform deploy applications in the namespace, they can select images in the public registry space for deployment through the image selector.
Sharing private images¶
Assuming that you have created a Managed Harbor or integrated to an external harbor, follow the steps below to share a private image with the namespace under the specified workspace (tenant):
The prerequisites are:
graph TB
create[Workspace created] --> bind[Namespace and workspace binded]
classDef plain fill:#ddd,stroke:#fff,stroke-width:1px,color:#000;
classDef k8s fill:#326ce5,stroke:#fff,stroke-width:1px,color:#fff;
classDef cluster fill:#fff,stroke:#bbb,stroke-width:1px,color:#326ce5;
class create,bind cluster;
click create "https://docs.daocloud.io/en/ghippo/user-guide/workspace/workspace/"
click bind "https://docs.daocloud.io/en/ghippo/user-guide/workspace/quota/"
Concrete steps are as follows: graph TB
create[Create managed Harbor] --> setpublic[Create a registry space and set it to public]
--> push[Push image]
--> bind[Registry space binding workspace]
--> deploy[Deploy application]
classDef plain fill:#ddd,stroke:#fff,stroke-width:1px,color:#000;
classDef k8s fill:#326ce5,stroke:#fff,stroke-width:1px,color:#fff;
classDef cluster fill:#fff,stroke:#bbb,stroke-width:1px,color:#326ce5;
class create,setpublic,push,bind,deploy cluster;
click create "https://docs.daocloud.io/en/kangaroo/hosted/"
click setpublic "https://docs.daocloud.io/en/kangaroo/integrate/create-space/"
click push "https://docs.daocloud.io/en/kangaroo/quickstart/push/"
click bind "https://docs.daocloud.io/en/kangaroo/integrate/integrate-admin/bind-to-ws/"
click deploy "https://docs.daocloud.io/en/kpanda/user-guide/workloads/create-deployment/"
Expected Result: Only when deploying applications in namespaces under this workspace, can you use the image selector to select private images under this registry space to deploy applications.
Tip
- The connected Harbor can achieve the same effect as above.
- Docker Registry itself only has public images, so after accessing, the images will be open to all namespaces.