Skip to content

ConfigMap creation methods

ConfigMap is a Kubernetes object used to decploy configuration artifacts from image content, allowing for the separation of code and configuration.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] It functions as a centralized storage for one or more configuration files or key-value pairs, which can be injected into containers at runtime.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] This abstraction reduces code coupling, as the same container image can be deployed across different environments (e.g., Development, Production) simply by swapping the associated ConfigMap.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md]

There are several primary methods to create a ConfigMap, ranging from imperative CLI commands to declarative YAML manifests.

1. Importing a file from the CLI

You can create a ConfigMap directly from an existing file using the kubectl create configmap command with the --from-file flag.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] This is useful for injecting configuration scripts or properties files.

Syntax: 

[kubectl](<./kubectl.md>) create [ConfigMap](<./configmap.md>) <map-name> --from-file=<filename>

Example: Creating a map named pg-initsql from a file named initdb.sql.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] 

[kubectl](<./kubectl.md>) create [ConfigMap](<./configmap.md>) pg-initsql --from-file=initdb.sql

2. Creating Key-Value pairs from the CLI

Using the --from-literal flag allows you to define specific key-value pairs without needing a source file.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] This method is ideal for simple environment variables or connection strings.

Syntax: 

[kubectl](<./kubectl.md>) create [ConfigMap](<./configmap.md>) <map-name> --from-literal=<key>=<value>

Example: Creating a map named pg-connect with database host and port information.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] 

[kubectl](<./kubectl.md>) create [ConfigMap](<./configmap.md>) pg-connect \
--from-literal=host=127.0.0.1 \
--from-literal=port=5432

3. Defining files via YAML

For a declarative Infrastructure as Code approach, you can define a ConfigMap in a YAML manifest.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] This method uses the data field to store file content, often utilizing a pipe symbol (|) to handle multi-line strings while preserving formatting.

Example: Creating initdb-yaml with an embedded SQL script.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] 

apiVersion: v1
kind: [ConfigMap](<./configmap.md>)
metadata:
  name: initdb-yaml
  labels:
    app: db
data:
  initdb.sql: |
    DROP TABLE IF EXISTS posts CASCADE;
    CREATE TABLE posts ( ... );

Once defined, the resource is applied using: 

[kubectl](<./kubectl.md>) apply -f initdb-configmap.yaml

4. Defining Key-Value pairs via YAML

Similar to the CLI literal method, YAML allows you to explicitly list key-value pairs under the data section.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md]

Example: Creating initdb-kv-yaml with user credentials.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md] 

apiVersion: v1
kind: [ConfigMap](<./configmap.md>)
metadata:
  name: initdb-kv-yaml
  labels:
    app: db
data:
  PG_USER: postgres
  PG_PASSWORD: postgres

Application in Pods

After creation, ConfigMaps are consumed by Pods primarily in two ways: 1. Environment Variables: Injecting specific keys as environment variables using configMapKeyRef. 2. Mounted Volumes: Mounting the entire ConfigMap or specific keys as files at a specific path (e.g., /docker-entrypoint-initdb.d) using the volumes and volumeMounts fields.^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md]

Sources

  • ^[400-devops-06-kubernetes-k8s-ithelp-day18-readme.md]