https://github.com/netricsag/tenant-api
API to present data to the tenant-dashboard
https://github.com/netricsag/tenant-api
capsule kubernetes tenants
Last synced: 7 months ago
JSON representation
API to present data to the tenant-dashboard
- Host: GitHub
- URL: https://github.com/netricsag/tenant-api
- Owner: netricsag
- License: apache-2.0
- Created: 2022-01-14T18:12:53.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-03-07T01:24:29.000Z (over 2 years ago)
- Last Synced: 2024-06-21T11:00:09.165Z (over 1 year ago)
- Topics: capsule, kubernetes, tenants
- Language: Go
- Homepage:
- Size: 432 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 7
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Codeowners: CODEOWNERS
- Security: SECURITY.md
Awesome Lists containing this project
README
# tenant-api
[](https://www.codefactor.io/repository/github/natron-io/tenant-api)
[](https://goreportcard.com/report/github.com/natron-io/tenant-api)
API to present data to the [tenant-dashboard](https://github.com/natron-io/tenant-dashboard) with a GitHub oauth login.
**Tenants** represents the **teams** of a GitHub organization.
## how it works
The tenant-api will search for namespaces named like the github teams, which you have access in your GitHub organisation.
It is recommended to use a multitenancy tool to jail each tenant in its host-Cluster namespace. For this you can use the [vclusters](https://vlcuster.com) technology. So you can deploy for each tenant a hostcluster namespace (named like your GitHub team) and in this namespace you can deploy the vcluster (which is the tenant). The vcluster will sync all resources created in it only on the hostcluster namespace. So the tenant-api only have to search the low level / costly resources (like pods, pvcs, ingress, requests, etc.) to present the data to the dashboard.
You can also sync your slack broadcast channel to present some important informations about your infrastructure to your tenant.
## api
#### `GET`
> **important:** for authenticated access you need to provide the `Authorization` header with the `Bearer` token.
You can add `` in front of the path to get the tenant specific data (of everything).
> e.g. `/api/v1//pods`
#### auth
`/login/github` - Login with GitHub \
`/login/github/callback` - Callback after GitHub login
#### notifications
`/api/v1/notifications` - Get the Slack notification messages of the broadcast channel provided via envs
##### general tenant resources
`/api/v1//pods` - Get a list of pods of a tenant \
`/api/v1//pvcs` - Get a list of pvcs of a tenant by storage classes \
`/api/v1//ingresses` - Get a list of ingresses of a tenant
##### specific tenant resources
`/api/v1//requests/cpu` - Get cpurequests in **Milicores** of a tenant \
`/api/v1//requests/memory` - Get memoryrequests in **Bytes** of a tenant \
`/api/v1//requests/storage` - Get storagerequests in **Bytes** of a tenant by storageclass \
##### tenant resources costs
`/api/v1//costs/cpu` - Get the CPU costs by CPU \
`/api/v1//costs/memory` - Get the memory costs by Memory \
`/api/v1//costs/storage` - Get the storage costs by StorageClass \
`/api/v1//costs/ingress` - Get the ingress costs by tenant
##### tenant resource quotas
`/api/v1//quotas/cpu` - Get the CPU resource Quota by the label defined via env \
`/api/v1//quotas/memory` - Get the memory resource Quota by the label defined via env \
`/api/v1//quotas/storage` - Get the storage resource Quota for each storage class by the label**s** defined via env
#### `POST`
##### auth
You can send the github code with json body `{"github_code": "..."}` to the `/login/github` endpoint.
> The code you need to generate must have the `read:org` scope.
## env
### general
`CORS` - CORS middleware for Fiber that that can be used to enable Cross-Origin Resource Sharing with various options. (e.g. "https://example.com, https://example2.com")
### GitHub
> There are two ways for authenticating with GitHub. You can authenticate without a dashboard, so the github callback url is not the same as the dashboard.
`CLIENT_ID` - GitHub client id **required** \
`CLIENT_SECRET` - GitHub client secret **required** \
`CALLBACK_URL` - GitHub oauth callback url without path *optional* (default: "http://localhost:3000")
### auth
`SECRET_KEY` - JWT secret key *optional* (default: random 32 bytes, displayed in the logs)
### api settings
`CORS` - Define CORS as one string *optional* (default: "*")
`MAX_REQUESTS` - Define max API requests per 30 Seconds *optional* (default: "100")
### notifications
`SLACK_TOKEN` - Tenant API Slack Application User Token *optional* (if not set, the notification REST route will be deactivated) \
`SLACK_BROADCAST_CHANNEL_ID` - BroadCast Slack Channel ID *optional* (**required** if SLACK_TOKEN is set) \
`SLACK_URL` - The slack url of your slack Channel *optional* (**required** if SLACK_TOKEN is set, e.g. "https://natronio.slack.com")
### cost calculation values
`DISCOUNT_LABEL` - label key for selecting the discount value *optional* (default: "natron.io/discount" (float -> e.g. "0.1")) \
`CPU_COST` - Cost of a CPU in your currency *optional* (default: 1.00 for 1 CPU) \
`MEMORY_COST` - Cost of a memory in your currency *optional* (default: 1.00 for 1 GB) \
`STORAGE_COST_` - Cost of your storage classes in your currency **required, multiple allowed** (default: 1.00 for 1 GB) \
`INGRESS_COST` - Cost of ingress in your currency *optional* (default: 1.00 for 1 ingress) \
`INGRESS_COST_PER_DOMAIN` - Calculates only ingress per domain.tld format *optional* (default: false) \
`EXCLUDE_INGRESS_VCLUSTER` - Excludes the vcluster ingress resource to expose the vcluster Kubernetes API. Name of the ingress must contain the string "vcluster" *optional* (default: false)
### resource quotas
It will get the resource quotas defined in the tenant namespace with the exact name of the tenant.
## labels
### resource quotas
For setting the quota labels on the tenant config namespace, you have to enter the values in the following format:
- CPU: `cores` e.g. natron.io/cpu-quota: "1" (-> 1 Core)
- Memory: `GB` e.g. natron.io/memory-quota: "4" (-> 4GB)
- Storage: `GB` e.g. natron.io/storage-quota-: "50" (-> 50GB)
## deployment
*example deployment files:* [kubernetes manifests](docs/kubernetes)
1. run a local minikube and apply a service account with clusterwide `view` permissions
```bash
minikube start
kubectl apply -f sa.yaml
kubectl apply -f rbac.yaml
kubectl apply -f deployment.yaml
kubectl expose deployment tenant-api --type=NodePort --port=8000
minikube service tenant-api
```
2. create a GitHub application in your GitHub organization and set the url (and port displayed at exposing the service via minikube) to the `CALLBACK_URL` (without path) and for the callback URL set the `CALLBACK_URL` with the path `/login/github/callback` (e.g. http://localhost:3000/login/github/callback)