The Kubernetes Integration is used to connect Shippable DevOps Assembly Lines platform to self-hosted Kubernetes so that you can deploy Docker based applications.
Adding account integration
Since this integration has been deprecated, you cannot create new account integrations for this, you can only edit/delete the exisiting Kubernetes integrations. You can use the new Kubernetes instead.
Creating a kubeconfig file
The best way to create an isolated set of credentials for use with Shippable Assembly Lines is to create a Kubernetes Service Account, and set up a custom kubeconfig file that utilizes it.
Create the Service Account
- Make sure you're on a machine that has a configured kubectl that can interact with your cluster
- Create a
shippable-service-account.yamlfile to represent the service account
apiVersion: v1 kind: ServiceAccount metadata: name: shippable-deploy #any name you'd like
- Use kubectl to create the service account on the master
$ kubectl create -f shippable-service-account.yaml serviceaccount "shippable-deploy" created
Create the kubeconfig file
First, we'll get the credentials for the service account that we created, then we'll add those credentials to a "context" within our kubeconfig file.
- Use kubectl to describe the service account so you can see its details
$ kubectl describe serviceAccounts shippable-deploy Name: shippable-deploy Namespace: default Labels: <none> Annotations: <none> Image pull secrets: <none> Mountable secrets: shippable-deploy-token-h6pdj Tokens: shippable-deploy-token-h6pdj
- Now describe the secret token associated with the account
$ kubectl describe secrets shippable-deploy-token-h6pdj Name: shippable-deploy-token-h6pdj Namespace: default Labels: <none> Annotations: kubernetes.io/service-account.name=shippable-deploy kubernetes.io/service-account.uid=da401cc4-3430-11e7-8529-42010a800fc8 Type: kubernetes.io/service-account-token Data ==== ca.crt: 1119 bytes namespace: 7 bytes token: eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJrdWJlcm5ldGVzL3Nlcn...
- Next, copy the existing kubeconfig from kubectl to a file so that it can be modified
$ kubectl config view --flatten --minify > myConfig.config
- Update the kubeconfig file to utilize the serviceAccount token. It should look similar to this:
apiVersion: v1 kind: Config users: - name: shippable-deploy user: token: eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJrdWJlcm5ldGVzL3Nlcn... clusters: - name: my-kube-cluster cluster: server: https://us-central1.sample.com certificate-authority-data: LS0tLS1CRUdJTiBDRVJUSUZJQ0FURS0tLS0tCk1JSUREREND... contexts: - context: cluster: my-kube-cluster user: shippable-deploy name: shippable-context current-context: shippable-context
- Make sure the
useris the name of the serviceAccount, and check that it is updated in the
userssection as well as the
- Check that the
current-contextstates the name of the context that references the correct cluster and the correct serviceAccount user.
- This assumes that the kubeconfig you started with already had the server and certificate authority information in the cluster section.
- Finally, take this kubeconfig file that you created, and use it to create your Shippable Kubernetes Integration.
Usage in Assembly Lines
The Kubernetes integration can be used in the following resources:
Default Environment Variables
When you create a resource with this integration, and use it as an
OUT for a
runCI job, a set of environment variables is automatically made available that you can use in your scripts.
<NAME> is the the friendly name of the resource with all letters capitalized and all characters that are not letters, numbers or underscores removed. For example,
my-key-1 will be converted to
my_key_1 will be converted to
||Name supplied in the integration|
||Access Type supplied in the integration|
||Kube config when kube master was chosen as access type|
||Kube config when bastion was chosen as access type|
||Bastion Host IP when bastion was chosen as access type|
||Private Key to access the bastion host|
||Public Key to access the bastion host|
Shippable Utility Functions
To make it easy to use these environment variables, the platform provides a command line utility that can be used to work with these values.
How to use these utility functions is documented here.