1
0
Fork 0
mirror of https://github.com/kyverno/kyverno.git synced 2025-03-09 17:37:12 +00:00
kyverno/docs/user/cli/commands/kyverno_apply.md
Ammar Yasser 739e6a21c4
Mutate existing CLI support (#11453)
* feat: Add flags for target resources and add fake client initialization

Signed-off-by: aerosouund <aerosound161@gmail.com>

* feat: Add fake discovery client and cluster bool in the policy processor

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Use the full mutation engine policy response in the engine response

Signed-off-by: aerosouund <aerosound161@gmail.com>

* feat: Extract mutated targets from the policy responses and print them out

Signed-off-by: aerosouund <aerosound161@gmail.com>

* feat: Add TargetResources field in the cli test schema

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Generate CLI crds

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: modify checkResult to take an arbitrary actual resource and resource name

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: change getAndCompareResource to take a resource name and cascade it to GetResourceFromPath

Signed-off-by: aerosouund <aerosound161@gmail.com>

* test: Create a simple test to test mutate existing in the CLI

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Allow GetResourceFromPath to select a resource with a name from a multi resource yaml

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Modify the runTest command to return the TestResponse type

- Create a fake client, load the target resources in it and use it in the PolicyProcessor.
- Create the TestResponse which contains Trigger and Target fields, each is a map of gvk/name to the responses corresponding to that resource.

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Rewrite output.go to use the TestResponse type

- Check for both target and trigger
- Create logic for appending the resource array in case no resources are passed
- Move row creation logic into a separate method to avoid code duplication
- Extract the proper target resource based on rule type
- Create a function to extract mutated target from the engine response

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Move tests to the correct folder

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Use apiVersion/Kind/Name as the key in the test responses

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Use the apiVersion/Kind/name key schema in checking results and fix invalid resource name checking for generate policies

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Use better variable names for rows

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Account for Generate resources being an array

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Use generated resource name in checking the results and printing output

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Uncomment checks printing

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Remove bug in engine response creation

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Move the generate logic into an else block

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Fix namespace fetching bug in cel validator

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Use pointer to int in the test counter

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Remove redundant method

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Skip resources not being found in the manifests

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Create another field in the engine to denote if this is a cluster engine or an offline engine

Simply checking for the client being nil is no longer enough because for cli operations the client will be a fake client
A pointer to bool is chosen because callers who don't necessarily know what to pass should be able to pass nil

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: Add extra argument in fake client initiation

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: add extra argument in fuzz test

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: Add extra arg

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: Handle resources specified as ns/name as this schema will be deprecated in favor of apiVersion/Kind/Name

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: Fix linter complaints

Signed-off-by: ammar <ammar.yasser@vodafone.com>

* fix: Use comma separation as array separators as kubernetes names don't support commas

To avoid undefined array length on splitting on /
using commas will result in a fixed length since all resources will have an apiVersion, kind, namespace and name

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Change resource array type to an array of any instead of array of string

To support the use of a string or a TestResourceSpec

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Expect the resource array to be an array of string or array of TestResourceSpec

Assert that an array element is either of these types and match the resources in both cases according to the element type
Expect that the key in responses is now separated by commas instead of slashes

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Expect that the resource array is now of type array of any and modify tests that use it

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Skip response check if the policy name isnt whats in the result

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Match the name if its specified as ns/name

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Fix linter complaint

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Run codegen

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Create CLI CRDs

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Run codegen

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Fix linter complaints

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Cleanup invalid code used in FixTest to adapt it to the schema changes

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Check if resource is nil before extracting

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: use the loadResources method to open targets in a directory

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Account for target resources with the same name but different namespaces

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Add CLI test for mutate existing with the same name

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Infer resource name and namespace from the actual resource and account for resources with the same name and namespace but different kinds

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: remove extra line

Signed-off-by: aerosouund <aerosound161@gmail.com>

* feat: Add printing mutate existing resources to the output or to a file

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Minor fixes

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: fix linter complaint

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: codegen

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Revert result back to error

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Use io discard to not print resources in the test command

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Update vague comments and remove outdated ones

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Integrate mutate existing changes with diff generation

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Move resource key generation into a function

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Add a mutate existing test that fails

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: fix linter complaint

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Remove redundant comment

Signed-off-by: aerosouund <aerosound161@gmail.com>

* refactor: Fix array of any assignment in cli test

Signed-off-by: aerosouund <aerosound161@gmail.com>

* fix: Dont check duplicate strings for field that is an array of any

Signed-off-by: aerosouund <aerosound161@gmail.com>

* bug: Fix appending to the wrong array

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: run fix tests

Signed-off-by: aerosouund <aerosound161@gmail.com>

* chore: Run fix tests

Signed-off-by: aerosouund <aerosound161@gmail.com>

---------

Signed-off-by: aerosouund <aerosound161@gmail.com>
Signed-off-by: ammar <ammar.yasser@vodafone.com>
Signed-off-by: Ammar Yasser <aerosound161@gmail.com>
Co-authored-by: ammar <ammar.yasser@vodafone.com>
Co-authored-by: shuting <shuting@nirmata.com>
2024-12-19 07:42:54 +00:00

5.8 KiB

kyverno apply

Applies policies on resources.

Synopsis

Applies policies on resources.

For more information visit https://kyverno.io/docs/kyverno-cli/#apply

kyverno apply [flags]

Examples

  # Apply on a resource
  kyverno apply /path/to/policy.yaml /path/to/folderOfPolicies --resource=/path/to/resource1 --resource=/path/to/resource2

  # Apply on a folder of resources
  kyverno apply /path/to/policy.yaml /path/to/folderOfPolicies --resource=/path/to/resources/

  # Apply on a cluster
  kyverno apply /path/to/policy.yaml /path/to/folderOfPolicies --cluster

  # Apply policies from a gitSourceURL on a cluster
  kyverno apply https://github.com/kyverno/policies/openshift/ --git-branch main --cluster

  # Apply single policy with variable on single resource
  kyverno apply /path/to/policy.yaml --resource /path/to/resource.yaml --set <variable1>=<value1>,<variable2>=<value2>

  # Apply multiple policy with variable on multiple resource
  kyverno apply /path/to/policy1.yaml /path/to/policy2.yaml --resource /path/to/resource1.yaml --resource /path/to/resource2.yaml -f /path/to/value.yaml

Options

      --audit-warn                         If set to true, will flag audit policies as warnings instead of failures
  -c, --cluster                            Checks if policies should be applied to cluster in the current context
      --context string                     The name of the kubeconfig context to use
      --continue-on-fail                   If set to true, will continue to apply policies on the next resource upon failure to apply to the current resource instead of exiting out
      --detailed-results                   If set to true, display detailed results
  -e, --exception strings                  Policy exception to be considered when evaluating policies against resources
      --exceptions strings                 Policy exception to be considered when evaluating policies against resources
      --exceptions-with-resources          Evaluate policy exceptions from the resources path
      --generate-exceptions                Generate policy exceptions for each violation
      --generated-exception-ttl duration   Default TTL for generated exceptions (default 720h0m0s)
  -b, --git-branch string                  test git repository branch
  -h, --help                               help for apply
      --kubeconfig string                  path to kubeconfig file with authorization and master location information
  -n, --namespace string                   Optional Policy parameter passed with cluster flag
  -o, --output string                      Prints the mutated/generated resources in provided file/directory
  -p, --policy-report                      Generates policy report when passed (default policyviolation)
      --registry                           If set to true, access the image registry using local docker credentials to populate external data
      --remove-color                       Remove any color from output
  -r, --resource strings                   Path to resource files
      --resources strings                  Path to resource files
  -s, --set strings                        Variables that are required
  -i, --stdin                              Optional mutate policy parameter to pipe directly through to kubectl
  -t, --table                              Show results in table format
      --target-resource strings            Path to individual files containing target resources files for policies that have mutate existing
      --target-resources strings           Path to a directory containing target resources files for policies that have mutate existing
  -u, --userinfo string                    Admission Info including Roles, Cluster Roles and Subjects
  -f, --values-file string                 File containing values for policy variables
      --warn-exit-code int                 Set the exit code for warnings; if failures or errors are found, will exit 1
      --warn-no-pass                       Specify if warning exit code should be raised if no objects satisfied a policy; can be used together with --warn-exit-code flag

Options inherited from parent commands

      --add_dir_header                   If true, adds the file directory to the header of the log messages
      --alsologtostderr                  log to standard error as well as files (no effect when -logtostderr=true)
      --log_backtrace_at traceLocation   when logging hits line file:N, emit a stack trace (default :0)
      --log_dir string                   If non-empty, write log files in this directory (no effect when -logtostderr=true)
      --log_file string                  If non-empty, use this log file (no effect when -logtostderr=true)
      --log_file_max_size uint           Defines the maximum size a log file can grow to (no effect when -logtostderr=true). Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
      --logtostderr                      log to standard error instead of files (default true)
      --one_output                       If true, only write logs to their native severity level (vs also writing to each lower severity level; no effect when -logtostderr=true)
      --skip_headers                     If true, avoid header prefixes in the log messages
      --skip_log_headers                 If true, avoid headers when opening log files (no effect when -logtostderr=true)
      --stderrthreshold severity         logs at or above this threshold go to stderr when writing to files and stderr (no effect when -logtostderr=true or -alsologtostderr=true) (default 2)
  -v, --v Level                          number for the log level verbosity
      --vmodule moduleSpec               comma-separated list of pattern=N settings for file-filtered logging

SEE ALSO

  • kyverno - Kubernetes Native Policy Management.