A helm v3 plugin to adopt existing k8s resources into a new generated helm chart

Last update: Dec 15, 2022

Comments: 3

helm-adopt

Overview

helm-adopt is a helm plugin to adopt existing k8s resources into a new generated helm chart, the idea behind the plugin was inspired by issue 2730.

the adopt plugin allows you to :

adopt existing k8s resources by generating a helm chart
migrate adopted resources to be controlled by helm
create a helm release using secret as driver
perform a dry-run/debug-mode (optional)

Note: adopt does not re-create resources.

Getting started

Installation

To install the plugin:

$ helm plugin install https://github.com/HamzaZo/helm-adopt

Update to latest

$ helm plugin update adopt

Install a specific version

$ helm plugin install https://github.com/HamzaZo/helm-adopt --version 0.1.0

You can also verify it's been installed using

$ helm plugin list

Usage

$ helm adopt
Adopt k8s resources into a new helm chart. It's expected to match plural resources kinds.

Usage:
  adopt [command]

Available Commands:
  help        Help about any command
  resources   adopt k8s resources into a new generated helm chart

Flags:
  -h, --help   help for adopt

Use "adopt [command] --help" for more information about a command.

Adopt k8s resources

$ helm adopt resources -h
Adopt k8s resources into a new generated helm chart 

Examples:
        
    $ helm adopt resources deployments:nginx services:my-svc -o/--output frontend

    $ helm adopt resources deployments:nginx clusterrolebindings:binding-rbac -o/--output frontend -n/--namespace <ns>

    $ helm adopt resources statefulsets:nginx services:my-svc -r/--release RELEASE-NAME -o/--output frontend -c/--kube-context <ctx>

    $ helm adopt resources deployments:nginx services:my-svc -r/--release RELEASE-NAME -o/--output frontend -k/--kubeconfig <kcfg>

Usage:
  adopt resources <pluralKind>:<name> [flags]

Flags:
      --debug-mode            Show the generated manifests on STDOUT
      --dry-run               Print what resources will be adopted 
  -h, --help                  help for resources
  -c, --kube-context string   name of the kubeconfig context to use
  -k, --kubeconfig string     path to the kubeconfig file
  -n, --namespace string      namespace scope for this request
  -o, --output string         Specify the chart directory of loaded yaml files
  -r, --release string        Specify the name for the generated release

Example User case

In the following example the adopt plugin will :

Generate a chart directory called example
adopt frontend,backend deployments and generate deployments-0.yaml deployments-1.yaml files.
adopt frontend,backend services and generate services-0.yaml services-1.yaml files.
adopt default-lr limitranges and generate limitranges-0.yaml file.
migrate adopted resources control to Helm
create example release with version 1 using secret as driver.

$ helm adopt resources deployments:frontend,backend services:frontend,backend limitranges:default-lr --output example --release example
INFO[0000] Adopting resources..                         
INFO[0000] Generating chart example                     
INFO[0000] Added resource as file deployments-0 into example chart 
INFO[0000] Added resource as file deployments-1 into example chart 
INFO[0000] Added resource as file limitranges-0 into example chart 
INFO[0000] Added resource as file services-0 into example chart 
INFO[0000] Added resource as file services-1 into example chart 
INFO[0000] Chart example is released as example.1

Note: You must use a semicolon after each name <pluralKind>:<name>,<name>,<name> if you need to adopt a bunch of resources under a specific pluralKind as shown in the example above.

check the release status

$ helm ls
NAME    NAMESPACE       REVISION        UPDATED                                 STATUS          CHART           APP VERSION
example default         1               2021-05-18 12:04:53.171943 +0200 CEST   deployed        example-0.1.0   0.1.0

$ kubectl get secret
NAME                            TYPE                                  DATA   AGE
sh.helm.release.v1.example.v1   helm.sh/release.v1                    1      51s

check generated chart directory.

$ tree example 
example
├── Chart.yaml
├── charts
├── templates
│   ├── _helpers.tpl
│   ├── deployments-0.yaml
│   ├── deployments-1.yaml
│   ├── limitranges-0.yaml
│   ├── services-0.yaml
│   └── services-1.yaml
└── values.yaml

2 directories, 8 files

Afterwards, you can edit the manifests in example/templates and turn them into templates and/or rework the generated chart into subcharts and then do a helm upgrade to fully transform the release.

Hints: To find out the pluralKind for a specific kind use kubectl api-resources

Similar Resources

The k8s-generic-webhook is a library to simplify the implementation of webhooks for arbitrary customer resources (CR) in the operator-sdk or controller-runtime.

k8s-generic-webhook The k8s-generic-webhook is a library to simplify the implementation of webhooks for arbitrary customer resources (CR) in the opera

Nov 24, 2022

Comments

Multiple matches for plural api resources causes issues

When pointing to for example 'networkpolicies' in the helm adopt command, its matching against the first one it finds, however many of my objects use the second api group instead.

networkpolicies crd.projectcalico.org true NetworkPolicy networkpolicies netpol networking.k8s.io true NetworkPolicy

I think the helm adopt command needs to possibly allow to (or make it mandatory) to specific explicitly the api group, in the case above I would give 'networkpolicies:networking.k8s.io:object1,object2', or something equivalent.
Not currently possible to adopt objects into an existing release

Is it possible to allow the possibility to adopt resources from one old release in to an already existing, new release without an error, before that new release is upgraded with a new chart containing that service.

eg a service moving from one release to another. I guess this use case might benefit greatly from helm adopt being able to gather its resource list from a rendered target chart/ helm template yaml. For me a great use case to fulfill would be to hook into the existing helm install/upgrade commands, and by passing some argument --adopt-objects we could adopt objects where we previously would have seen an error that the objects already exist.

good job 2233333333323333333

$ helm adopt resources gateways:nginx-gateway virtualservices:nginx-virtual  --output test --release test
INFO[0000] Adopting resources..
INFO[0000] Generating chart test
INFO[0000] Added resource as file gateways-0 into test chart
INFO[0000] Added resource as file virtualservices-0 into test chart
INFO[0000] Chart test is released as test.1

$ cat templates/gateways-0.yaml
apiVersion: networking.istio.io/v1beta1
kind: Gateway
metadata:
  name: nginx-gateway
  namespace: default
spec:
  selector:
    istio: ingressgateway
  servers:
  - hosts:
    - nginx.google.com
    port:
      name: https
      number: 443
      protocol: HTTPS
    tls:
      credentialName: secret-tls
      mode: SIMPLE
$ cat templates/virtualservices-0.yaml
apiVersion: networking.istio.io/v1beta1
kind: VirtualService
metadata:
  name: nginx-virtual
  namespace: default
spec:
  gateways:
  - nginx-gateway
  hosts:
  - nginx.google.com
  http:
  - route:
    - destination:
        host: nginx.default.svc.cluster.local
        port:
          number: 80

A helm v3 plugin to adopt existing k8s resources into a new generated helm chart

helm-adopt

Overview

Getting started

Installation

Usage

Adopt k8s resources

Example User case

Owner

Hamza ZOUHAIR

Similar Resources

The k8s-generic-webhook is a library to simplify the implementation of webhooks for arbitrary customer resources (CR) in the operator-sdk or controller-runtime.

Poc rsa - A simple golang scaffolding to help me to create new api projects or workers with golang on k8s

A helm v3 plugin to get values from a previous release

Plugin for Helm to integrate the sigstore ecosystem

Application open new tab in chrome when your favourite youtuber add new video.

CoreDNS plugin implementing K8s multi-cluster services DNS spec.

K8s-socketcan - Virtual SocketCAN Kubernetes device plugin

A kubectl plugin for easier query and operate k8s cluster.

A kubernetes plugin which enables dynamically add or remove GPU resources for a running Pod

Comments

Multiple matches for plural api resources causes issues

Not currently possible to adopt objects into an existing release

good job 2233333333323333333

Related tags

Katenary - Convert docker-compose to a configurable helm chart

K8s-cinder-csi-plugin - K8s Pod Use Openstack Cinder Volume

A k8s vault webhook is a Kubernetes webhook that can inject secrets into Kubernetes resources by connecting to multiple secret managers

A tool to bring existing Azure resources under Terraform's management

A tool to bring existing Azure resources under Terraform's management

Helm Operator is designed to managed the full lifecycle of Helm charts with Kubernetes CRD resource.

helm-lint-ls is helm lint language server protocol LSP.

Write controller-runtime based k8s controllers that read/write to git, not k8s

K8s-ingress-health-bot - A K8s Ingress Health Bot is a lightweight application to check the health of the ingress endpoints for a given kubernetes namespace.

K8s-go-structs - All k8s API Go structs