K8s Dashboard banner

K8s Dashboard

14 devlogs
81h 43m 31s

It is a self-hosted Platform-as-a-Service (PaaS) designed to bring the developer experience of Vercel or Cloud Run to your own hardware. Built on top of K3s, it turns any cheap Linux server into a powerful serverless engine. It features instant au…

It is a self-hosted Platform-as-a-Service (PaaS) designed to bring the developer experience of Vercel or Cloud Run to your own hardware. Built on top of K3s, it turns any cheap Linux server into a powerful serverless engine. It features instant automatic HTTPS, persistent storage management, and a unique “Scale-to-Zero” (only for HTTP) architecture that hibernates your apps when they aren’t in use—saving resources and money.

This project uses AI

He helped me at readme and debug typescript headache ( also Golang stressful)

Demo Repository

Loading README...

nbth

Add registering, and I hope that everyone can consider my note before reject
Also, the demo video here demo video

Attachment
0
nbth

What news

  • We have a new validation check on the route
  • The video demo of the dashboard

Upcoming

  • inter-cluster networking
0
nbth

i found the problem why sometimes a user cannot login

Attachment
0
nbth

What news

  • the implementation of volume mount in the pod and deployment
  • the implementation for StorageClass, Persistent Volumes, and Persistent Volume Claims
  • env from some of the k8s metadata that the pod can access, like podid,nodeid, and other
  • adding debug container and refactor pod show ui for easier in choosing which container
  • add event getting ( tracking k8s action )

Upcoming

  • the implmentation of statefull set
Attachment
Attachment
Attachment
Attachment
Attachment
0
nbth

What new

  • now deployment can list the pod inside it
  • bulk env add for the configmap, secret and other where that have the env input
  • to many deployment bug that i found
  • re-deployment endpoint will be add in next ui commit
  • permission checking for ref editor
  • dynamic new permission logic handler for frontend ( ownership principle for permission )
  • cleanup auto-created resource if that not present in k8s cluster

Ongoing

  • inter-cluster networking, serect, configmap pvc
  • add describe for event checking for example use want to know why pod die or why deployment dont run they will use this to debug
  • api endpoint to allow user get raw k8s object for user check object config
  • add rollout option and also subcommand of them
    – [ ] rollout history
    – [-] rollout pause ( next week plan )
    – [x] rollout restart ( implemented at /deployments/:cluster-id/re-deploy/:id )
    – [-] rollout resume ( next week plan )
    – [-] rollout status ( will add when we have the way to manage demonset )
    – [ ] rollout undo
  • add statefullset resource management for
  • add kubectl debug <podID> -it --copy-to=debug-pod --container=<containerName> -- sh and kubectl attach debug-pod -c mindconnect -i -t
Attachment
Attachment
0
nbth

What new:

  • The full functionality of secret and configmap works
  • The PVC to allow Traefik to share the acme.json together
  • new style of these page ( pod, deployment, etc….)
  • try the demo at nginx.k8s.nbth.dev ( don’t take the ddos attack please )

Upcoming:

  • global configmap and secret
  • the most valuable thing like i will deploy mindconnect to here
  • add new choose to allow user to choose does they want to expose port as http or https or both
  • Re-deployment button
Attachment
Attachment
Attachment
Attachment
Attachment
0
nbth

What’s new:

  • the new navbar and sidebar
  • The new ssl auto-renew feature
  • agent auto-update via GitHub Release

Upcoming:

  • config map and secret for more backend not covered by k8s
  • the demo is going out
  • migrate to nginx ???
  • mcp for agent
Attachment
0
nbth

What news:

  • Introduce a new Secret and ConfigMap page for management, and also k8s integration is included.

Upcoming:

  • They can link to pod and deployment ( yeah )
Attachment
Attachment
0
nbth

Introducing the brand new pod ( and deployment ) port expose ( this runs best with HTTP, but TCP and UDP may cause resource consumption problems)
Note: You also need to set up a record in your DNS settings
Background task for next devlog:

  • The brand new “Scale to Zero” for deployment
Attachment
Attachment
Attachment
Attachment
Attachment
0
nbth

Now you can get the pod log and the exec command in the pod

  • behind the scenes: I am currently developing pod expose and will be stable in future
Attachment
Attachment
0
nbth

now you can create and delete pods

  • behind the screen : i have done service and deployment manager but that still have some problem
  • the logs have some auth problem so you cannot watch log now
Attachment
Attachment
Attachment
0
nbth

they send me heartbeat
now server now the state of k8s cluster

Attachment
0
nbth

hello we have the cluster creation now. yeah
Behind the screen: at the backend, we have the api for pod and node manager
at the agent we almost done CURD for pod and node in k8s
the work for next update:
complete the agent resource heartbeat for sync k8s state to database

Attachment
Attachment
0
nbth

starting with the login page
Behind the scenes: we have the agent with the basics of development CURD

Attachment
Attachment
0