Chapter 12. Configuring functions

12.1. Accessing secrets and config maps from functions using CLI

After your functions have been deployed to the cluster, they can access data stored in secrets and config maps. This data can be mounted as volumes, or assigned to environment variables. You can configure this access interactively by using the Knative CLI, or by manually by editing the function configuration YAML file.

Important

To access secrets and config maps, the function must be deployed on the cluster. This functionality is not available to a function running locally.

If a secret or config map value cannot be accessed, the deployment fails with an error message specifying the inaccessible values.

12.1.1. Modifying function access to secrets and config maps interactively

You can manage the secrets and config maps accessed by your function by using the kn func config interactive utility. The available operations include listing, adding, and removing values stored in config maps and secrets as environment variables, as well as listing, adding, and removing volumes. This functionality enables you to manage what data stored on the cluster is accessible by your function.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Run the following command in the function project directory:

    $ kn func config

    Alternatively, you can specify the function project directory using the --path or -p option.

  2. Use the interactive interface to perform the necessary operation. For example, using the utility to list configured volumes produces an output similar to this:

    $ kn func config
    ? What do you want to configure? Volumes
    ? What operation do you want to perform? List
    Configured Volumes mounts:
    - Secret "mysecret" mounted at path: "/workspace/secret"
    - Secret "mysecret2" mounted at path: "/workspace/secret2"

    This scheme shows all operations available in the interactive utility and how to navigate to them:

    kn func config
       ├─> Environment variables
       │               ├─> Add
       │               │    ├─> ConfigMap: Add all key-value pairs from a config map
       │               │    ├─> ConfigMap: Add value from a key in a config map
       │               │    ├─> Secret: Add all key-value pairs from a secret
       │               │    └─> Secret: Add value from a key in a secret
       │               ├─> List: List all configured environment variables
       │               └─> Remove: Remove a configured environment variable
       └─> Volumes
               ├─> Add
               │    ├─> ConfigMap: Mount a config map as a volume
               │    └─> Secret: Mount a secret as a volume
               ├─> List: List all configured volumes
               └─> Remove: Remove a configured volume
  3. Optional. Deploy the function to make the changes take effect:

    $ kn func deploy -p test

12.1.2. Modifying function access to secrets and config maps interactively by using specialized commands

Every time you run the kn func config utility, you need to navigate the entire dialogue to select the operation you need, as shown in the previous section. To save steps, you can directly execute a specific operation by running a more specific form of the kn func config command:

  • To list configured environment variables:

    $ kn func config envs [-p <function-project-path>]
  • To add environment variables to the function configuration:

    $ kn func config envs add [-p <function-project-path>]
  • To remove environment variables from the function configuration:

    $ kn func config envs remove [-p <function-project-path>]
  • To list configured volumes:

    $ kn func config volumes [-p <function-project-path>]
  • To add a volume to the function configuration:

    $ kn func config volumes add [-p <function-project-path>]
  • To remove a volume from the function configuration:

    $ kn func config volumes remove [-p <function-project-path>]

12.2. Configuring your function project using the func.yaml file

The func.yaml file contains the configuration for your function project. Values specified in func.yaml are used when you execute a kn func command. For example, when you run the kn func build command, the value in the build field is used. In some cases, you can override these values with command line flags or environment variables.

12.2.1. Referencing local environment variables from func.yaml fields

If you want to avoid storing sensitive information such as an API key in the function configuration, you can add a reference to an environment variable available in the local environment. You can do this by modifying the envs field in the func.yaml file.

Prerequisites

  • You need to have the function project created.
  • The local environment needs to contain the variable that you want to reference.

Procedure

  • To refer to a local environment variable, use the following syntax:

    {{ env:ENV_VAR }}

    Substitute ENV_VAR with the name of the variable in the local environment that you want to use.

    For example, you might have the API_KEY variable available in the local environment. You can assign its value to the MY_API_KEY variable, which you can then directly use within your function:

    Example function

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - name: MY_API_KEY
      value: '{{ env:API_KEY }}'
    ...

12.2.2. Adding annotations to functions

You can add Kubernetes annotations to a deployed Serverless function. Annotations enable you to attach arbitrary metadata to a function, for example, a note about the function’s purpose. Annotations are added to the annotations section of the func.yaml configuration file.

There are two limitations of the function annotation feature:

  • After a function annotation propagates to the corresponding Knative service on the cluster, it cannot be removed from the service by deleting it from the func.yaml file. You must remove the annotation from the Knative service by modifying the YAML file of the service directly, or by using the OpenShift Container Platform web console.
  • You cannot set annotations that are set by Knative, for example, the autoscaling annotations.

12.2.3. Adding annotations to a function

You can add annotations to a function. Similar to a label, an annotation is defined as a key-value map. Annotations are useful, for example, for providing metadata about a function, such as the function’s author.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For every annotation that you want to add, add the following YAML to the annotations section:

    name: test
    namespace: ""
    runtime: go
    ...
    annotations:
      <annotation_name>: "<annotation_value>" 1
    1
    Substitute <annotation_name>: "<annotation_value>" with your annotation.

    For example, to indicate that a function was authored by Alice, you might include the following annotation:

    name: test
    namespace: ""
    runtime: go
    ...
    annotations:
      author: "alice@example.com"
  3. Save the configuration.

The next time you deploy your function to the cluster, the annotations are added to the corresponding Knative service.

12.2.4. Additional resources

12.2.5. Adding function access to secrets and config maps manually

You can manually add configuration for accessing secrets and config maps to your function. This might be preferable to using the kn func config interactive utility and commands, for example when you have an existing configuration snippet.

12.2.5.1. Mounting a secret as a volume

You can mount a secret as a volume. Once a secret is mounted, you can access it from the function as a regular file. This enables you to store on the cluster data needed by the function, for example, a list of URIs that need to be accessed by the function.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For each secret you want to mount as a volume, add the following YAML to the volumes section:

    name: test
    namespace: ""
    runtime: go
    ...
    volumes:
    - secret: mysecret
      path: /workspace/secret
    • Substitute mysecret with the name of the target secret.
    • Substitute /workspace/secret with the path where you want to mount the secret.

      For example, to mount the addresses secret, use the following YAML:

      name: test
      namespace: ""
      runtime: go
      ...
      volumes:
      - configMap: addresses
        path: /workspace/secret-addresses
  3. Save the configuration.

12.2.5.2. Mounting a config map as a volume

You can mount a config map as a volume. Once a config map is mounted, you can access it from the function as a regular file. This enables you to store on the cluster data needed by the function, for example, a list of URIs that need to be accessed by the function.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For each config map you want to mount as a volume, add the following YAML to the volumes section:

    name: test
    namespace: ""
    runtime: go
    ...
    volumes:
    - configMap: myconfigmap
      path: /workspace/configmap
    • Substitute myconfigmap with the name of the target config map.
    • Substitute /workspace/configmap with the path where you want to mount the config map.

      For example, to mount the addresses config map, use the following YAML:

      name: test
      namespace: ""
      runtime: go
      ...
      volumes:
      - configMap: addresses
        path: /workspace/configmap-addresses
  3. Save the configuration.

12.2.5.3. Setting environment variable from a key value defined in a secret

You can set an environment variable from a key value defined as a secret. A value previously stored in a secret can then be accessed as an environment variable by the function at runtime. This can be useful for getting access to a value stored in a secret, such as the ID of a user.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For each value from a secret key-value pair that you want to assign to an environment variable, add the following YAML to the envs section:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - name: EXAMPLE
      value: '{{ secret:mysecret:key }}'
    • Substitute EXAMPLE with the name of the environment variable.
    • Substitute mysecret with the name of the target secret.
    • Substitute key with the key mapped to the target value.

      For example, to access the user ID that is stored in userdetailssecret, use the following YAML:

      name: test
      namespace: ""
      runtime: go
      ...
      envs:
      - value: '{{ configMap:userdetailssecret:userid }}'
  3. Save the configuration.

12.2.5.4. Setting environment variable from a key value defined in a config map

You can set an environment variable from a key value defined as a config map. A value previously stored in a config map can then be accessed as an environment variable by the function at runtime. This can be useful for getting access to a value stored in a config map, such as the ID of a user.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For each value from a config map key-value pair that you want to assign to an environment variable, add the following YAML to the envs section:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - name: EXAMPLE
      value: '{{ configMap:myconfigmap:key }}'
    • Substitute EXAMPLE with the name of the environment variable.
    • Substitute myconfigmap with the name of the target config map.
    • Substitute key with the key mapped to the target value.

      For example, to access the user ID that is stored in userdetailsmap, use the following YAML:

      name: test
      namespace: ""
      runtime: go
      ...
      envs:
      - value: '{{ configMap:userdetailsmap:userid }}'
  3. Save the configuration.

12.2.5.5. Setting environment variables from all values defined in a secret

You can set an environment variable from all values defined in a secret. Values previously stored in a secret can then be accessed as environment variables by the function at runtime. This can be useful for simultaneously getting access to a collection of values stored in a secret, for example, a set of data pertaining to a user.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For every secret for which you want to import all key-value pairs as environment variables, add the following YAML to the envs section:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - value: '{{ secret:mysecret }}' 1
    1
    Substitute mysecret with the name of the target secret.

    For example, to access all user data that is stored in userdetailssecret, use the following YAML:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - value: '{{ configMap:userdetailssecret }}'
  3. Save the configuration.

12.2.5.6. Setting environment variables from all values defined in a config map

You can set an environment variable from all values defined in a config map. Values previously stored in a config map can then be accessed as environment variables by the function at runtime. This can be useful for simultaneously getting access to a collection of values stored in a config map, for example, a set of data pertaining to a user.

Prerequisites

  • The OpenShift Serverless Operator and Knative Serving are installed on the cluster.
  • You have installed the Knative (kn) CLI.
  • You have created a function.

Procedure

  1. Open the func.yaml file for your function.
  2. For every config map for which you want to import all key-value pairs as environment variables, add the following YAML to the envs section:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - value: '{{ configMap:myconfigmap }}' 1
    1
    Substitute myconfigmap with the name of the target config map.

    For example, to access all user data that is stored in userdetailsmap, use the following YAML:

    name: test
    namespace: ""
    runtime: go
    ...
    envs:
    - value: '{{ configMap:userdetailsmap }}'
  3. Save the file.

12.3. Configurable fields in func.yaml

You can configure some of the func.yaml fields.

12.3.1. Configurable fields in func.yaml

Many of the fields in func.yaml are generated automatically when you create, build, and deploy your function. However, there are also fields that you modify manually to change things, such as the function name or the image name.

12.3.1.1. buildEnvs

The buildEnvs field enables you to set environment variables to be available to the environment that builds your function. Unlike variables set using envs, a variable set using buildEnv is not available during function runtime.

You can set a buildEnv variable directly from a value. In the following example, the buildEnv variable named EXAMPLE1 is directly assigned the one value:

buildEnvs:
- name: EXAMPLE1
  value: one

You can also set a buildEnv variable from a local environment variable. In the following example, the buildEnv variable named EXAMPLE2 is assigned the value of the LOCAL_ENV_VAR local environment variable:

buildEnvs:
- name: EXAMPLE1
  value: '{{ env:LOCAL_ENV_VAR }}'

12.3.1.2. envs

The envs field enables you to set environment variables to be available to your function at runtime. You can set an environment variable in several different ways:

  1. Directly from a value.
  2. From a value assigned to a local environment variable. See the section "Referencing local environment variables from func.yaml fields" for more information.
  3. From a key-value pair stored in a secret or config map.
  4. You can also import all key-value pairs stored in a secret or config map, with keys used as names of the created environment variables.

This examples demonstrates the different ways to set an environment variable:

name: test
namespace: ""
runtime: go
...
envs:
- name: EXAMPLE1 1
  value: value
- name: EXAMPLE2 2
  value: '{{ env:LOCAL_ENV_VALUE }}'
- name: EXAMPLE3 3
  value: '{{ secret:mysecret:key }}'
- name: EXAMPLE4 4
  value: '{{ configMap:myconfigmap:key }}'
- value: '{{ secret:mysecret2 }}' 5
- value: '{{ configMap:myconfigmap2 }}' 6
1
An environment variable set directly from a value.
2
An environment variable set from a value assigned to a local environment variable.
3
An environment variable assigned from a key-value pair stored in a secret.
4
An environment variable assigned from a key-value pair stored in a config map.
5
A set of environment variables imported from key-value pairs of a secret.
6
A set of environment variables imported from key-value pairs of a config map.

12.3.1.3. builder

The builder field specifies the strategy used by the function to build the image. It accepts values of pack or s2i.

12.3.1.4. build

The build field indicates how the function should be built. The value local indicates that the function is built locally on your machine. The value git indicates that the function is built on a cluster by using the values specified in the git field.

12.3.1.5. volumes

The volumes field enables you to mount secrets and config maps as a volume accessible to the function at the specified path, as shown in the following example:

name: test
namespace: ""
runtime: go
...
volumes:
- secret: mysecret 1
  path: /workspace/secret
- configMap: myconfigmap 2
  path: /workspace/configmap
1
The mysecret secret is mounted as a volume residing at /workspace/secret.
2
The myconfigmap config map is mounted as a volume residing at /workspace/configmap.

12.3.1.6. options

The options field enables you to modify Knative Service properties for the deployed function, such as autoscaling. If these options are not set, the default ones are used.

These options are available:

  • scale

    • min: The minimum number of replicas. Must be a non-negative integer. The default is 0.
    • max: The maximum number of replicas. Must be a non-negative integer. The default is 0, which means no limit.
    • metric: Defines which metric type is watched by the Autoscaler. It can be set to concurrency, which is the default, or rps.
    • target: Recommendation for when to scale up based on the number of concurrently incoming requests. The target option can be a float value greater than 0.01. The default is 100, unless the options.resources.limits.concurrency is set, in which case target defaults to its value.
    • utilization: Percentage of concurrent requests utilization allowed before scaling up. It can be a float value between 1 and 100. The default is 70.
  • resources

    • requests

      • cpu: A CPU resource request for the container with deployed function.
      • memory: A memory resource request for the container with deployed function.
    • limits

      • cpu: A CPU resource limit for the container with deployed function.
      • memory: A memory resource limit for the container with deployed function.
      • concurrency: Hard Limit of concurrent requests to be processed by a single replica. It can be integer value greater than or equal to 0, default is 0 - meaning no limit.

This is an example configuration of the scale options:

name: test
namespace: ""
runtime: go
...
options:
  scale:
    min: 0
    max: 10
    metric: concurrency
    target: 75
    utilization: 75
  resources:
    requests:
      cpu: 100m
      memory: 128Mi
    limits:
      cpu: 1000m
      memory: 256Mi
      concurrency: 100

12.3.1.7. image

The image field sets the image name for your function after it has been built. You can modify this field. If you do, the next time you run kn func build or kn func deploy, the function image will be created with the new name.

12.3.1.8. imageDigest

The imageDigest field contains the SHA256 hash of the image manifest when the function is deployed. Do not modify this value.

12.3.1.9. labels

The labels field enables you to set labels on a deployed function.

You can set a label directly from a value. In the following example, the label with the role key is directly assigned the value of backend:

labels:
- key: role
  value: backend

You can also set a label from a local environment variable. In the following example, the label with the author key is assigned the value of the USER local environment variable:

labels:
- key: author
  value: '{{ env:USER }}'

12.3.1.10. name

The name field defines the name of your function. This value is used as the name of your Knative service when it is deployed. You can change this field to rename the function on subsequent deployments.

12.3.1.11. namespace

The namespace field specifies the namespace in which your function is deployed.

12.3.1.12. runtime

The runtime field specifies the language runtime for your function, for example, python.