Skip to content

Latest commit

 

History

History
115 lines (102 loc) · 5.46 KB

kubeconfig.adoc

File metadata and controls

115 lines (102 loc) · 5.46 KB

.kubeconfig files

Overview

In order to easily switch between multiple clusters, a .kubeconfig file was defined. This file contains a series of authentication mechanisms and cluster connection information associated with nicknames. It also introduces the concept of a tuple of authentication information (user) and cluster connection information called a context that is also associated with a nickname.

Multiple files are .kubeconfig files are allowed. At runtime they are loaded and merged together along with override options specified from the command line (see rules below).

Manipulation of .kubeconfig via openshift ex config <subcommand>

In order to more easily manipulate .kubeconfig files, there are a series of subcommands to openshift ex config to help.

openshift ex config set-credentials name --auth-path=path/to/authfile --client-certificate=path/to/cert --client-key=path/to/key --token=string
  Sets a user entry in .kubeconfig.  If the referenced name already exists, the specified information will be merged in.
openshift ex config set-cluster name --server=server --skip-tls=bool --certificate-authority=path/to/ca --api-version=string
  Sets a cluster entry in .kubeconfig.  If the referenced name already exists, the specified information will be merged in.
openshift ex config set-context name --user=string --cluster=string --namespace=string
  Sets a config entry in .kubeconfig.  If the referenced name already exists, the specified information will be merged in.
openshift ex config use-context name
  Sets current-context to name
openshift ex config set property-name property-value
  Sets arbitrary value in .kubeconfig
openshift ex config unset property-name
  Unsets arbitrary value in .kubeconfig
openshift ex config view --local=true --global=false --kubeconfig=specific/filename --merged
  Displays the merged (or not) result of the specified .kubeconfig file

--local, --global, and --kubeconfig are valid flags for all of these operations.

Example

$openshift ex config set-credentials myself --auth-path=path/to/my/existing/auth-file
$openshift ex config set-cluster local-server --server=http://localhost:8080
$openshift ex config set-context default-context --cluster=local-server --user=myself
$openshift ex config use-context default-context
$openshift ex config set contexts.default-context.namespace the-right-prefix
$openshift ex config view

produces this output

clusters:
  local-server:
    server: http://localhost:8080
contexts:
  default-context:
    cluster: local-server
    namespace: the-right-prefix
    user: myself
current-context: default-context
preferences: {}
users:
  myself:
    auth-path: path/to/my/existing/auth-file

and a .kubeconfig file that looks like this

apiVersion: v1
clusters:
- cluster:
    server: http://localhost:8080
  name: local-server
contexts:
- context:
    cluster: local-server
    namespace: the-right-prefix
    user: myself
  name: default-context
current-context: default-context
kind: Config
preferences: {}
users:
- name: myself
  user:
    auth-path: path/to/my/existing/auth-file

Loading and merging rules

The rules for loading and merging the .kubeconfig files are straightforward, but there are a lot of them. The final config is built in this order:

  1. Merge together the kubeconfig itself. This is done with the following hierarchy and merge rules: Empty filenames are ignored. Files with non-deserializable content produced errors. The first file to set a particular value or map key wins and the value or map key is never changed. This means that the first file to set CurrentContext will have its context preserved. It also means that if two files specify a "red-user", only values from the first file’s red-user are used. Even non-conflicting entries from the second file’s "red-user" are discarded.

    1. CommandLineLocation - the value of the kubeconfig command line option

    2. EnvVarLocation - the value of $KUBECONFIG

    3. CurrentDirectoryLocation - pwd/.kubeconfig

    4. HomeDirectoryLocation = ~/.kube/.kubeconfig

  2. Determine the context to use based on the first hit in this chain

    1. command line argument - the value of the context command line option

    2. current-context from the merged kubeconfig file

    3. Empty is allowed at this stage

  3. Determine the cluster info and user to use. At this point, we may or may not have a context. They are built based on the first hit in this chain. (run it twice, once for user, once for cluster)

    1. command line argument - user for user name and cluster for cluster name

    2. If context is present, then use the context’s value

    3. Empty is allowed

  4. Determine the actual cluster info to use. At this point, we may or may not have a cluster info. Build each piece of the cluster info based on the chain (first hit wins):

    1. command line arguments - server, api-version, certificate-authority, and insecure-skip-tls-verify

    2. If cluster info is present and a value for the attribute is present, use it.

    3. If you don’t have a server location, error.

  5. User is build using the same rules as cluster info, EXCEPT that you can only have one authentication technique per user. The command line flags are: auth-path, client-certificate, client-key, and token. If there are two conflicting techniques, fail.

  6. For any information still missing, use default values and potentially prompt for authentication information