repository·master·Indexed 23 days ago
https://github.com/tkestack/tkeAn open-source container management platform for production-grade Kubernetes deployments. TKEStack provides unified cluster management, workload orchestration, and advanced O&M capabilities via a web console and CLI. The documentation includes guides for integrating Ceph storage using the ceph-csi-cephfs and ceph-csi-rbd Helm charts.
ceph-csi-cephfs chart is used to add CephFS volume support to a Kubernetes cluster, enabling the use of CephFS as a storage backend for persistent volumes.TKEStack is a container management platform designed for production environments. Key capabilities include:
tke-installer component, refer to the Introduction for tke-installer.The Harbor Preheat API allows for managing P2P (Peer-to-Peer) provider instances and preheat policies within Harbor projects. All URIs are relative to http://localhost/api/v2.0.
| Method | HTTP Request | Description |
|---|---|---|
| CreateInstance | Post /p2p/preheat/instances | Create p2p provider instances |
| CreatePolicy | Post /projects/{project_name}/preheat/policies | Create a preheat policy under a project |
| DeleteInstance | Delete /p2p/preheat/instances/{preheat_instance_name} | Delete the specified P2P provider instance |
| DeletePolicy | Delete /projects/{project_name}/preheat/policies/{preheat_policy_name} | Delete a preheat policy |
| GetExecution | Get /projects/{project_name}/preheat/policies/{preheat_policy_name}/executions/{execution_id} | Get an execution detail by id |
| GetInstance | Get /p2p/preheat/instances/{preheat_instance_name} | Get a P2P provider instance |
| GetPolicy | Get /projects/{project_name}/preheat/policies/{preheat_policy_name} | Get a preheat policy |
| GetPreheatLog | Get /projects/{project_name}/preheat/policies/{preheat_policy_name}/executions/{execution_id}/tasks/{task_id}/logs | Get the log text stream of the specified task |
| ListExecutions | Get /projects/{project_name}/preheat/policies/{preheat_policy_name}/executions | List executions for the given policy |
| ListInstances | Get /p2p/preheat/instances | List P2P provider instances |
| ListPolicies | Get /projects/{project_name}/preheat/policies | List preheat policies |
| ListProviders | Get /p2p/preheat/providers | List P2P providers |
| ListProvidersUnderProject | Get /projects/{project_name}/preheat/providers | Get all providers at project level |
| ListTasks | Get /projects/{project_name}/preheat/policies/{preheat_policy_name}/executions/{execution_id}/tasks | List all related tasks for an execution |
| ManualPreheat | Post /projects/{project_name}/preheat/policies/{preheat_policy_name} | Manual preheat |
| PingInstances | Post /p2p/preheat/instances/ping | Ping status of an instance |
| StopExecution | Patch /projects/{project_name}/preheat/policies/{preheat_policy_name}/executions/{execution_id} | Stop an execution |
| UpdateInstance | Put /p2p/preheat/instances/{preheat_instance_name} | Update the specified P2P provider instance |
| UpdatePolicy | Put /projects/{project_name}/preheat/policies/{preheat_policy_name} | Update preheat policy |
The Harbor Artifact API provides endpoints for managing artifacts within projects and repositories. All URIs are relative to http://localhost/api/v2.0. Common operations include adding/removing labels, creating/deleting tags, copying artifacts, and retrieving artifact details.
Note on Repository Names: If a repositoryName contains a slash (e.g., a/b), it must be URL encoded (e.g., a%252Fb).
ceph-csi-rbd chart provides RBD (Rados Block Device) volume support for your Kubernetes cluster. This allows you to provision and use Ceph RBD volumes as persistent storage.ProjectApi provides a set of methods to manage Harbor projects, including creating, deleting, listing, and retrieving project details or logs. All URIs are relative to http://localhost/api/v2.0. Most methods require a context.Context for authentication and tracing, and use basic authorization.Kubernetes events record cluster operations, resource scheduling, and lifecycle changes. By default, Kubernetes only retains these events in ETCD for approximately one hour.
PersistentEvent is a TKEStack feature that enables real-time export of Kubernetes cluster events to a specified ElasticSearch index for long-term storage and troubleshooting.
When installed, the following object is deployed to your cluster:
| Object Name | Type | Default Resources | Namespace |
|---|---|---|---|
tke-persistent-event | deployment | 0.2 CPU, 100MB RAM | kube-system |
Cron Horizontal Pod Autoscaler (CronHPA) allows users to perform scheduled automatic scaling for supported workload objects (such as Deployment, StatefulSet, and TApp) using CronTab syntax.
It works by introducing a new Custom Resource Definition (CRD) called cronhpas.extensions.tkestack.io. The cron-hpa-controller (an operator) monitors this CRD and parses the configuration to execute scaling operations based on the system time.
TApp is a custom Kubernetes Workload type (based on CRDs) designed for complex, non-microservice applications that standard Kubernetes workloads like Deployment or StatefulSet cannot easily manage.
Key Use Cases:
Conceptually, TApp ≈ Deployment + StatefulSet + Job, providing a unified workload type that supports both service-oriented and batch-oriented tasks.
LogAgent is a TKEStack add-on that provides cluster-wide log collection. It replaces the older LogCollector component. Once enabled, it runs as a DaemonSet in the kube-system namespace to collect logs and send them to consumers like Kafka or Elasticsearch.
Supported Collection Types:
Stderr and Stdout from containers. Logs are output in JSON format with Kubernetes metadata (labels, annotations, etc.).Resource Usage (per node):