CLAMP participants (kubernetes, http) Smoke Tests

1. Introduction

The CLAMP participants (kubernetes and http) are used to interact with the helm client in a kubernetes environment for the deployment of microservices via helm chart as well as to configure the microservices over REST endpoints. Both of these participants are often used together in the Automation Composition Management workflow.

This document will serve as a guide to do smoke tests on the different components that are involved when working with the participants and outline how they operate. It will also show a developer how to set up their environment for carrying out smoke tests on these participants.

2. Setup Guide

This article assumes that:

  • You are using the operating systems such as linux/macOS/windows.

  • You are using a directory called git off your home directory (~/git) for your git repositories

  • Your local maven repository is in the location ~/.m2/repository

  • You have copied the settings.xml from oparent to ~/.m2/ directory

  • You have added settings to access the ONAP Nexus to your M2 configuration, see Maven Settings Example (bottom of the linked page)

  • Your local helm is in the location /usr/local/bin/helm

  • Your local kubectl is in the location /usr/local/bin/kubectl

The procedure documented in this article has been verified using Ubuntu 20.04 LTS VM.

2.1 Prerequisites

2.2 Cloning CLAMP automation composition

Run the below command to clone the required CLAMP automation composition:

cd ~/git
git clone https://gerrit.onap.org/r/policy/clamp clamp

Execution of the command above results in the following directory hierarchy in your ~/git directory:

  • ~/git/clamp

2.3 Setting up the components

2.3.1 Setting for kubernetes.

If the helm location is not ‘/usr/local/bin/helm’ or the kubectl location is not ‘/usr/local/bin/kubectl’, you have to update the file ‘participant/participant-impl/participant-impl-kubernetes/src/main/java/org/onap/policy/clamp/acm/participant/kubernetes/helm/HelmClient.java’.

2.3.2 Building CLAMP automation composition

Step 1: Optionally, for a completely clean build, remove the ONAP built modules from your local repository.

rm -fr ~/.m2/repository/org/onap

Step 2: You can now build the Policy framework.

Build java artifacts and docker images:

cd ~/git/clamp
mvn clean install -P docker -DskipTests

2.3.3 Running Postgres and Kafka

We will be using Docker to run our Postgres instance and Kafka. It will have the acm-runtime database running in it. Create the db-pg.conf and db-pg.sh files in the directory ~/git.

# Copyright (C) 2022, 2024,2026 OpenInfra Foundation Europe. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#         http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

PGSQL_ROOT_PASSWORD=secret
PGSQL_USER=policy_user
PGSQL_PASSWORD=policy_user
PGPASSWORD=policy_user
POSTGRES_PASSWORD=policy_user
# Copyright (C) 2022, 2024,2026 OpenInfra Foundation Europe. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#         http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

psql -U postgres -d postgres --command "CREATE USER ${PGSQL_USER} WITH PASSWORD '${PGSQL_PASSWORD}';"

for db in migration policyadmin operationshistory clampacm
do
    psql -U postgres -d postgres --command "CREATE DATABASE ${db};"
    psql -U postgres -d postgres --command "ALTER DATABASE ${db} OWNER TO ${PGSQL_USER} ;"
    psql -U postgres -d postgres --command "GRANT ALL PRIVILEGES ON DATABASE ${db} TO ${PGSQL_USER} ;"
done

Create the ‘docker-compose.yaml’ using following code:

services:
  postgres:
    image: nexus3.onap.org:10001/library/postgres:latest
    container_name: postgres
    hostname: postgres
    command: [ '--idle_in_transaction_session_timeout=28800' ]
    env_file: ./db-pg.conf
    volumes:
      - ./db-pg.sh:/docker-entrypoint-initdb.d/db-pg.sh:ro
    expose:
      - 5432
    ports:
      - "5432:5432"

  zookeeper:
    image: nexus3.onap.org:10001/confluentinc/cp-zookeeper:latest
    environment:
      ZOOKEEPER_CLIENT_PORT: 2181
      ZOOKEEPER_TICK_TIME: 2000
    ports:
      - 2181:2181

  kafka:
    image: nexus3.onap.org:10001/confluentinc/cp-kafka:7.4.9
    container_name: kafka
    depends_on:
      - zookeeper
    ports:
      - 29092:29092
      - 9092:9092
    environment:
      KAFKA_BROKER_ID: 1
      KAFKA_ZOOKEEPER_CONNECT: zookeeper:2181
      KAFKA_ADVERTISED_LISTENERS: PLAINTEXT://kafka:9092,PLAINTEXT_HOST://localhost:29092
      KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: PLAINTEXT:PLAINTEXT,PLAINTEXT_HOST:PLAINTEXT
      KAFKA_INTER_BROKER_LISTENER_NAME: PLAINTEXT
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1

Run the docker composition:

cd ~/git/
docker compose up

2.3.4 Running ACM-R on the Command Line

To start the automation composition runtime service, we need to execute the following command line from the “runtime-acm” directory in the clamp repo. Automation composition runtime uses the config file “src/main/resources/application.yaml” by default.

cd ~/git/clamp/runtime-acm
java -DRUNTIME_USER=runtimeUser -DRUNTIME_PASSWORD=zb\!XztG34 \
     -DSQL_HOST=localhost -DSQL_PORT=5432 -DSQL_USER=policy_user -DSQL_PASSWORD=policy_user \
     -DKAFKA_SERVER=localhost:29092 -DTOPIC_COMM_INFRASTRUCTURE=kafka \
     -jar target/policy-clamp-runtime-acm-9.0.1-SNAPSHOT.jar

2.3.5 Helm chart repository

Kubernetes participant consumes helm charts from the local chart database as well as from a helm repository. For the smoke testing, we are going to add nginx-stable helm repository to the helm client. The following command can be used to add nginx repository to the helm client.

helm repo add nginx-stable https://helm.nginx.com/stable

2.3.6 Kubernetes and http participants

The participants can be started from the clamp repository by executing the following command line from the appropriate directories. The participants will be started and get registered to the Automation composition runtime.

Navigate to the directory “participant/participant-impl/participant-impl-kubernetes/” and start kubernetes participant.

cd ~/git/clamp/participant/participant-impl/participant-impl-kubernetes
java -Dserver.port=8082 -DkafkaServer=localhost:29092 -DtopicCommInfrastructure=kafka \
     -jar target/policy-clamp-participant-impl-kubernetes-9.0.1-SNAPSHOT.jar

Navigate to the directory “participant/participant-impl/participant-impl-http/” and start http participant.

cd ~/git/clamp/participant/participant-impl/participant-impl-http
java -Dserver.port=8083 -DkafkaServer=localhost:29092 -DtopicCommInfrastructure=kafka \
     -jar target/policy-clamp-participant-impl-http-9.0.1-SNAPSHOT.jar

3. Running Tests

In this section, we will run through the sequence of steps in ACM workflow . The workflow can be triggered via Postman client.

3.1 Commissioning

Commission Automation composition TOSCA definitions to Runtime.

The Automation composition definitions are commissioned to runtime-acm which populates the ACM runtime database. The following sample TOSCA template is commissioned to the runtime endpoint which contains definitions for kubernetes participant that deploys nginx ingress microservice helm chart and a http POST request for http participant.

Tosca Service Template

Commissioning Endpoint:

POST: https://<Runtime ACM IP> : <Port> /onap/policy/clamp/acm/v2/compositions

A successful commissioning gives 201 responses in the postman client.

3.2 Prime an Automation composition definition

Once the template is commissioned, we can prime it. This will connect AC definition with related participants.

Prime Endpoint:

PUT: https://<Runtime ACM IP> : <Port> /onap/policy/clamp/acm/v2/compositions/{compositionId}

Request body:

{
     "primeOrder": "PRIME"
}

A successful prime request gives 202 responses in the postman client.

3.3 Create New Instances of Automation composition

Once AC definition is primes, we can instantiate automation composition instances. This will create the instances with default state “UNDEPLOYED”.

Instantiation Endpoint:

POST: https://<Runtime ACM IP> : <Port> /onap/policy/clamp/acm/v2/compositions/{compositionId}/instances

Request body:

Instantiation json

A successful creation of new instance gives 201 responses in the postman client.

3.4 Change the State of the Instance

When the automation composition is updated with state “DEPLOYED”, the Kubernetes participant fetches the node template for all automation composition elements and deploys the helm chart of each AC element into the cluster. The following sample json input is passed on the request body.

Automation Composition Update Endpoint:

PUT: https://<Runtime ACM IP> : <Port> /onap/policy/clamp/acm/v2/compositions/{compositionId}/instances/{instanceId}

Request body:
{
 "deployOrder": "DEPLOY"
}

A successful deploy request gives 202 responses in the postman client. After the state changed to “DEPLOYED”, nginx-ingress pod is deployed in the kubernetes cluster. And http participant should have posted the dummy data to the configured URL in the tosca template.

The following command can be used to verify the pods deployed successfully by kubernetes participant.

helm ls -n onap | grep nginx
kubectl get po -n onap | grep nginx

The overall state of the automation composition should be “DEPLOYED” to indicate both the participants has successfully completed the operations. This can be verified by the following rest endpoint.

Verify automation composition state:

GET: https://<Runtime ACM IP> : <Port>/onap/policy/clamp/acm/v2/compositions/{compositionId}/instances/{instanceId}

3.5 Automation Compositions can be “UNDEPLOYED” after deployment

By changing the state to “UNDEPLOYED”, all the helm deployments under the corresponding automation composition will be uninstalled from the cluster. Automation Composition Update Endpoint:

PUT: https://<Runtime ACM IP> : <Port> /onap/policy/clamp/acm/v2/compositions/{compositionId}/instances/{instanceId}

Request body:
{
 "deployOrder": "UNDEPLOY"
}

The nginx pod should be deleted from the k8s cluster.

This concludes the required smoke tests for http and kubernetes participants.