Microservice Configuration Profiles

This tutorial describes three methods to utilize configuration profiles to specify resource allocation for Entando microservices. These provide a simple way to customize microservice deployment parameters for improved efficiency.

Prerequisites

• A working instance of Entando
• Verify dependencies with the Entando CLI: ent check-env develop
• Add an Entando Operator ConfigMap if needed, then
  • Enable this property under the data section so the Entando Operator can manage the resource settings:

   entando.k8s.operator.impose.limits: "true"
  

Profile Options

A profile is a set of configurations encoded as YAML embedded in the OperatorConfigMap as a string, since ConfigMaps cannot be multilevel.

The three methods to insert configuration profiles are:

1. Inline Profile
2. Mapped Profile
3. Default Profile

Configuration

Currently, the resources for memory and CPU can be specified. When you specify a limit on a resource, it is the upper limit allowed for the container. The units for the resources are listed below but should not be included in the specification. Resource settings help Kubernetes determine in which node a pod should be created.

resources.limits.cpu:       integer, millicpus
resources.limits.memory:    integer, mebibytes

The examples below use YOUR-PLUGIN, YOUR-PLUGIN-CODE, YOUR-ORG, YOUR-BUNDLE and YOUR-PROFILE-NAME as placeholders. Also note the use of |- to designate a new line in the code.

Retrieve the Plugin Code

You will need to retrieve the plugin code, which is calculated during installation and written to the EntandoPlugin custom resource as part of the deployment of the microservice.

Use the following command from the root bundle project directory, where YOUR-ORG is your Docker organization and YOUR-BUNDLE contains the microservice:

ent ecr get-plugin-code YOUR-ORG/YOUR-PLUGIN-NAME --repo=docker://registry.hub.docker.com/YOUR-ORG/YOUR-BUNDLE

Method 1: Inline Profile

Add the resource parameters to the OperatorConfigMap as an inline profile at data/entando.profile.plugins.YOUR-PLUGIN-CODE:

data:
  entando.profile.plugins.YOUR-PLUGIN-CODE: |-
    resources.limits.cpu: "1000"
    resources.limits.memory: "2000"

Method 2: Mapped Profile

1. Create the resource parameter profile in the OperatorConfigMap of the data profile at data/entando.profile.YOUR-PROFILE-NAME:

data:
  entando.profile.YOUR-PROFILE-NAME: |-
    resources.limits.cpu: "1000"
    resources.limits.memory: "2000"

2. Add a reference in the profileMapping file at data/entando.plugins.profileMapping:

data:
  entando.plugins.profileMapping: |-
    YOUR-PLUGIN-CODE: YOUR-PROFILE-NAME

Method 3: Default Profile

1. Add the resource parameter profile to the OperatorConfigMap in the data section at data/entando.profile.YOUR-PROFILE-NAME:

data:
  entando.profile.YOUR-PROFILE-NAME: |-
    resources.limits.cpu: "1000"
    resources.limits.memory: "2000"

2. Add a reference to the profile at data/entando.plugins.defaultProfile:

data:
  entando.plugins.defaultProfile: YOUR-PROFILE-NAME

Note: If a resourceRequirement was specified in the plugin custom resource, that will override a profile.

References

Please refer to the Kubernetes documentation on Resources (opens new window) for more details.