Skip to content

Latest commit

 

History

History
73 lines (57 loc) · 2.12 KB

README.md

File metadata and controls

73 lines (57 loc) · 2.12 KB

🐿️ Sanjab

GitHub release (with filter) GitHub top language GitHub Workflow Status (with event)

Sanjab (aka squirrel is english) is a service for backing up your k8s objects into Ceph cluster. Whenever a new request is sent to api-server, sanjab gets the kubernetes object of that request and stores it Ceph cluster.

Sanjab is built with kubernetes operator pattern. It watches over the following resources on cluster, and stores the object as a yaml file on Ceph cluster.

  • Pods
  • Deployments
  • Services
  • Cronjobs
  • Configmaps
  • Secrets
  • ServiceAccounts
  • StatefulSets
  • HPAs
  • Ingress
  • PVCs

🌲 configs

For Sanjab server configs, you need to create a config.yml file. This file's structure is like this:

timeout: 60 # in seconds
port: 80 # service http port
namespace: "default" # kubernetes namespace
ceph_disable: false # enabling ceph upload
ceph:
  host: "http://127.0.0.1:6800"
  access: "access-token"
  secret: "secret-token"
  bucket: "bucket-name"

You can list your desire resources in the config.yml file. Sanjab creates a worker process for each resource to manage them concurrently. For example, if we want to back up only pods and deployments, we need to set our resources like this:

resources:
  - pods
  - deployments

⛺ deployment

To deploy Sanjab on a kubernetes cluster, review deployment manifests. You need to deploy all of those files, especially rbac and service account. Sanjab needs access to all api groups in order to get, list, and watch those resources.

🥜 docker image

docker pull amirhossein21/sanjab:latest

🚨 monitoring

If you want to check Sanjab health and workers status, make the following http requests.

curl -i -X GET localhost/metrics # metrics export
curl -i -X GET localhost/workers # workers status
curl -i -X GET localhost/health  # service health