CLAMP Policy Participant Smoke Tests
1. Introduction
The Smoke testing of the policy participant is executed in a local CLAMP/Policy environment. The CLAMP-ACM interfaces interact with the Policy Framework to perform actions based on the state of the policy participant. The goal of the Smoke tests is the ensure that CLAMP Policy Participant and Policy Framework work together as expected. All applications will be running by console, so they need to run with different ports. Configuration files should be changed accordingly.
Application |
port |
---|---|
MariDB |
3306 |
Zookeeper |
2181 |
Kafka |
29092 |
policy-api |
6968 |
policy-pap |
6970 |
policy-clamp-runtime-acm |
6969 |
onap/policy-clamp-ac-pf-ppnt |
8085 |
2. Setup Guide
This section will show the developer how to set up their environment to start testing in GUI with some instruction on how to carry out the tests. There are several prerequisites. Note that this guide is written by a Linux user - although the majority of the steps show will be exactly the same in Windows or other systems.
2.1 Prerequisites
Java 17
Maven 3.9
Git
Refer to this guide for basic environment setup Setting up dev environment
2.2 Cloning CLAMP automation composition and all dependency
Run a script such as the script below to clone the required modules from the ONAP git repository. This script clones CLAMP automation composition and all dependency.
1 #!/usr/bin/env bash
2
3 ## script name for output
4 MOD_SCRIPT_NAME='basename $0'
5
6 ## the ONAP clone directory, defaults to "onap"
7 clone_dir="onap"
8
9 ## the ONAP repos to clone
10 onap_repos="\
11 policy/api \
12 policy/clamp \
13 policy/pap "
14
15 ##
16 ## Help screen and exit condition (i.e. too few arguments)
17 ##
18 Help()
19 {
20 echo ""
21 echo "$MOD_SCRIPT_NAME - clones all required ONAP git repositories"
22 echo ""
23 echo " Usage: $MOD_SCRIPT_NAME [-options]"
24 echo ""
25 echo " Options"
26 echo " -d - the ONAP clone directory, defaults to '.'"
27 echo " -h - this help screen"
28 echo ""
29 exit 255;
30 }
31
32 ##
33 ## read command line
34 ##
35 while [ $# -gt 0 ]
36 do
37 case $1 in
38 #-d ONAP clone directory
39 -d)
40 shift
41 if [ -z "$1" ]; then
42 echo "$MOD_SCRIPT_NAME: no clone directory"
43 exit 1
44 fi
45 clone_dir=$1
46 shift
47 ;;
48
49 #-h prints help and exists
50 -h)
51 Help;exit 0;;
52
53 *) echo "$MOD_SCRIPT_NAME: undefined CLI option - $1"; exit 255;;
54 esac
55 done
56
57 if [ -f "$clone_dir" ]; then
58 echo "$MOD_SCRIPT_NAME: requested clone directory '$clone_dir' exists as file"
59 exit 2
60 fi
61 if [ -d "$clone_dir" ]; then
62 echo "$MOD_SCRIPT_NAME: requested clone directory '$clone_dir' exists as directory"
63 exit 2
64 fi
65
66 mkdir $clone_dir
67 if [ $? != 0 ]
68 then
69 echo cannot clone ONAP repositories, could not create directory '"'$clone_dir'"'
70 exit 3
71 fi
72
73 for repo in $onap_repos
74 do
75 repoDir=`dirname "$repo"`
76 repoName=`basename "$repo"`
77
78 if [ ! -z $dirName ]
79 then
80 mkdir "$clone_dir/$repoDir"
81 if [ $? != 0 ]
82 then
83 echo cannot clone ONAP repositories, could not create directory '"'$clone_dir/repoDir'"'
84 exit 4
85 fi
86 fi
87
88 git clone https://gerrit.onap.org/r/${repo} $clone_dir/$repo
89 done
90
91 echo ONAP has been cloned into '"'$clone_dir'"'
Execution of the script above results in the following directory hierarchy in your ~/git directory:
~/git/onap
~/git/onap/policy
~/git/onap/policy/api
~/git/onap/policy/clamp
~/git/onap/policy/pap
2.3 Building CLAMP automation composition and all dependency
Step 1: Setting topicParameterGroup for kafka localhost in clamp and policy-participant. It needs to set ‘kafka’ as topicCommInfrastructure and ‘localhost:29092’ as server. In the clamp repo, you should find the file ‘runtime-acm/src/main/resources/application.yaml’. This file (in the ‘runtime’ parameters section) may need to be altered as below:
runtime:
topics:
operationTopic: policy-acruntime-participant
syncTopic: acm-ppnt-sync
participantParameters:
heartBeatMs: 20000
maxStatusWaitMs: 150000
maxOperationWaitMs: 200000
topicParameterGroup:
topicSources:
- topic: ${runtime.topics.operationTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
fetchTimeout: 15000
topicSinks:
- topic: ${runtime.topics.operationTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
- topic: ${runtime.topics.syncTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
acmParameters:
toscaElementName: org.onap.policy.clamp.acm.AutomationCompositionElement
toscaCompositionName: org.onap.policy.clamp.acm.AutomationComposition
Setting topicParameterGroup for kafka localhost and api/pap http client (in the ‘participant’ parameters section) may need to be apply into the file ‘participant/participant-impl/participant-impl-policy/src/main/resources/config/application.yaml’.
participant:
pdpGroup: defaultGroup
pdpType: apex
policyApiParameters:
clientName: api
hostname: localhost
port: 6968
userName: policyadmin
password: zb!XztG34
useHttps: false
allowSelfSignedCerts: false
policyPapParameters:
clientName: pap
hostname: localhost
port: 6970
userName: policyadmin
password: zb!XztG34
useHttps: false
allowSelfSignedCerts: false
intermediaryParameters:
topics:
operationTopic: policy-acruntime-participant
syncTopic: acm-ppnt-sync
reportingTimeIntervalMs: 120000
description: Participant Description
participantId: 101c62b3-8918-41b9-a747-d21eb79c6c03
clampAutomationCompositionTopics:
topicSources:
- topic: ${participant.intermediaryParameters.topics.operationTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
fetchTimeout: 15000
- topic: ${participant.intermediaryParameters.topics.syncTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
fetchTimeout: 15000
topicSinks:
- topic: ${participant.intermediaryParameters.topics.operationTopic}
servers:
- localhost:29092
topicCommInfrastructure: kafka
participantSupportedElementTypes:
-
typeName: org.onap.policy.clamp.acm.PolicyAutomationCompositionElement
typeVersion: 1.0.0
Step 2: Setting datasource.url, hibernate.ddl-auto and server.port in policy-api. In the api repo, you should find the file ‘main/src/main/resources/application.yaml’. This file may need to be altered as below:
spring:
profiles:
active: default
security.user:
name: policyadmin
password: zb!XztG34
mvc.converters.preferred-json-mapper: gson
datasource:
url: jdbc:mariadb://localhost:3306/policyadmin
driverClassName: org.mariadb.jdbc.Driver
username: policy_user
password: policy_user
jpa:
hibernate:
ddl-auto: none
naming:
physical-strategy: org.hibernate.boot.model.naming.PhysicalNamingStrategyStandardImpl
implicit-strategy: org.onap.policy.common.spring.utils.CustomImplicitNamingStrategy
server:
port: 6968
servlet:
context-path: /policy/api/v1
Step 3: Setting datasource.url, server.port, and api http client in policy-pap. In the pap repo, you should find the file ‘main/src/main/resources/application.yaml’. This file may need to be altered as below:
spring:
security:
user:
name: policyadmin
password: zb!XztG34
datasource:
url: jdbc:mariadb://localhost:3306/policyadmin
driverClassName: org.mariadb.jdbc.Driver
username: policy_user
password: policy_user
jpa:
hibernate:
ddl-auto: none
naming:
physical-strategy: org.hibernate.boot.model.naming.PhysicalNamingStrategyStandardImpl
implicit-strategy: org.onap.policy.common.spring.utils.CustomImplicitNamingStrategy
mvc:
converters:
preferred-json-mapper: gson
server:
port: 6970
servlet:
context-path: /policy/pap/v1
pap:
name: PapGroup
topic:
pdp-pap.name: POLICY-PDP-PAP
notification.name: POLICY-NOTIFICATION
heartbeat.name: POLICY-HEARTBEAT
pdpParameters:
heartBeatMs: 120000
updateParameters:
maxRetryCount: 1
maxWaitMs: 30000
stateChangeParameters:
maxRetryCount: 1
maxWaitMs: 30000
topicParameterGroup:
topicSources:
- topic: ${pap.topic.pdp-pap.name}
servers:
- kafka
topicCommInfrastructure: NOOP
fetchTimeout: 15000
- topic: ${pap.topic.heartbeat.name}
effectiveTopic: ${pap.topic.pdp-pap.name}
consumerGroup: policy-pap
servers:
- kafka
topicCommInfrastructure: NOOP
fetchTimeout: 15000
topicSinks:
- topic: ${pap.topic.pdp-pap.name}
servers:
- kafka
topicCommInfrastructure: NOOP
- topic: ${pap.topic.notification.name}
servers:
- kafka
topicCommInfrastructure: NOOP
healthCheckRestClientParameters:
- clientName: api
hostname: localhost
port: 6968
userName: policyadmin
password: zb!XztG34
useHttps: false
basePath: policy/api/v1/healthcheck
- clientName: distribution
hostname: policy-distribution
port: 6969
userName: healthcheck
password: zb!XztG34
useHttps: true
basePath: healthcheck
- clientName: kafka
hostname: kafka
port: 3905
useHttps: true
basePath: topics
management:
endpoints:
web:
base-path: /
exposure:
include: health, metrics, prometheus
path-mapping:
-metrics: plain-metrics
-prometheus: metrics
Step 4: Optionally, for a completely clean build, remove the ONAP built modules from your local repository.
rm -fr ~/.m2/repository/org/onap
Step 5: A pom such as the one below can be used to build the ONAP Policy Framework modules. Create the pom.xml file in the directory ~/git/onap/policy.
1 <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
2 <modelVersion>4.0.0</modelVersion>
3 <groupId>org.onap</groupId>
4 <artifactId>onap-policy</artifactId>
5 <version>1.0.0-SNAPSHOT</version>
6 <packaging>pom</packaging>
7 <name>${project.artifactId}</name>
8 <inceptionYear>2024</inceptionYear>
9 <organization>
10 <name>ONAP</name>
11 </organization>
12
13 <modules>
14 <module>api</module>
15 <module>clamp</module>
16 <module>pap</module>
17 </modules>
18 </project>
Step 6: You can now build the Policy framework.
Build java artifacts only:
cd ~/git/onap/policy mvn clean install -DskipTests
Build with docker images:
cd ~/git/onap/policy/clamp/packages/ mvn clean install -P docker cd ~/git/onap/policy/api/packages/ mvn clean install -P docker cd ~/git/onap/policy/pap/packages/ mvn clean install -P docker
2.4 Setting up the components
2.4.1 MariaDB and Kafka Setup
We will be using Docker to run our mariadb instance`and Zookeeper/Kafka. It will have a total of two databases running in mariadb.
clampacm: the policy-clamp-runtime-acm db
policyadmin: the policy-api db
Step 1: Create the mariadb.sql file in a directory ~/git.
create database clampacm; CREATE USER 'policy'@'%' IDENTIFIED BY 'P01icY'; GRANT ALL PRIVILEGES ON clampacm.* TO 'policy'@'%'; CREATE DATABASE `policyadmin`; CREATE USER 'policy_user'@'%' IDENTIFIED BY 'policy_user'; GRANT ALL PRIVILEGES ON policyadmin.* to 'policy_user'@'%'; CREATE DATABASE `migration`; GRANT ALL PRIVILEGES ON migration.* to 'policy_user'@'%'; FLUSH PRIVILEGES;
Step 2: Create the init.sh file in a directory ~/git with execution permission.
#!/bin/sh export POLICY_HOME=/opt/app/policy export SQL_USER=${MYSQL_USER} export SQL_PASSWORD=${MYSQL_PASSWORD} export SCRIPT_DIRECTORY=sql /opt/app/policy/bin/prepare_upgrade.sh ${SQL_DB} /opt/app/policy/bin/db-migrator -s ${SQL_DB} -o report /opt/app/policy/bin/db-migrator -s ${SQL_DB} -o upgrade rc=$? /opt/app/policy/bin/db-migrator -s ${SQL_DB} -o report nc -l -p 6824 exit $rc
Step 3: Create the wait_for_port.sh file in a directory ~/git with execution permission.
#!/bin/sh usage() { echo args: [-t timeout] [-c command] hostname1 port1 hostname2 port2 ... >&2 exit 1 } tmout=300 cmd= while getopts c:t: opt do case "$opt" in c) cmd="$OPTARG" ;; t) tmout="$OPTARG" ;; *) usage ;; esac done nargs=$((OPTIND-1)) shift "$nargs" even_args=$(($#%2)) if [ $# -lt 2 ] || [ "$even_args" -ne 0 ] then usage fi while [ $# -ge 2 ] do export host="$1" export port="$2" shift shift echo "Waiting for $host port $port..." while [ "$tmout" -gt 0 ] do if command -v docker > /dev/null 2>&1 then docker ps --format "table {{ .Names }}\t{{ .Status }}" fi nc -vz "$host" "$port" rc=$? if [ $rc -eq 0 ] then break else tmout=$((tmout-1)) sleep 1 fi done if [ $rc -ne 0 ] then echo "$host port $port cannot be reached" exit $rc fi done $cmd exit 0
Step 4: Create the ‘docker-compose.yaml’ using following code:
services:
mariadb:
image: mariadb:10.10.2
command: ['mysqld', '--lower_case_table_names=1']
volumes:
- type: bind
source: ./mariadb.sql
target: /docker-entrypoint-initdb.d/data.sql
environment:
- MYSQL_ROOT_PASSWORD=my-secret-pw
ports:
- "3306:3306"
policy-db-migrator:
image: nexus3.onap.org:10001/onap/policy-db-migrator:3.1.3-SNAPSHOT
container_name: policy-db-migrator
hostname: policy-db-migrator
depends_on:
- mariadb
expose:
- 6824
environment:
SQL_DB: policyadmin
SQL_HOST: mariadb
MYSQL_ROOT_PASSWORD: my-secret-pw
MYSQL_USER: policy_user
MYSQL_PASSWORD: policy_user
MYSQL_CMD: mysql
volumes:
- ./init.sh:/opt/app/policy/bin/db_migrator_policy_init.sh:ro
- ./wait_for_port.sh:/opt/app/policy/bin/wait_for_port.sh:ro
entrypoint: /opt/app/policy/bin/wait_for_port.sh
command: [
'-c',
'/opt/app/policy/bin/db_migrator_policy_init.sh',
'mariadb', '3306'
]
zookeeper:
image: confluentinc/cp-zookeeper:latest
environment:
ZOOKEEPER_CLIENT_PORT: 2181
ZOOKEEPER_TICK_TIME: 2000
ports:
- 2181:2181
kafka:
image: confluentinc/cp-kafka:latest
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
Step 5: Run the docker composition:
cd ~/git/ docker compose up
2.4.2 Policy API
In the policy-api repo, navigate to the “/main” directory. You can then run the following command to start the policy api:
mvn spring-boot:run
2.4.3 Policy PAP
In the policy-pap repo, navigate to the “/main” directory. You can then run the following command to start the policy pap:
mvn spring-boot:run
2.4.4 ACM Runtime
To start the clampacm runtime we need to go the “runtime-acm” directory in the clamp repo. You can then run the following command to start the clampacm runtime:
mvn spring-boot:run
2.4.5 ACM Policy Participant
To start the policy participant we need to go to the “participant/participant-impl/participant-impl-policy” directory in the clamp repo. You can then run the following command to start the policy-participant:
mvn spring-boot:run
3. Testing Procedure
3.1 Testing Outline
To perform the Smoke testing of the policy-participant we will be verifying the behaviours of the participant when the ACM changes state. The scenarios are:
UNDEPLOYED to DEPLOYED: participant creates policies and policyTypes specified in the ToscaServiceTemplate using policy-api and deploys the policies using pap.
LOCK to UNLOCK: participant changes lock state to UNLOCK. No operation performed.
UNLOCK to LOCK: participant changes lock state to LOCK. No operation performed.
DEPLOYED to UNDEPLOYED: participant undeploys deployed policies and deletes policies and policyTypes which have been created.
3.2 Testing Steps
Creation of AC Definition:
An AC Definition is created by commissioning a Tosca template. Using postman, commission a TOSCA template using the following template:
To verify this, we check that the AC Definition has been created and is in state COMMISSIONED.
Priming AC Definition:
The AC Definition state is changed from COMMISSIONED to PRIMED using postman:
{
"primeOrder": "PRIME"
}
To verify this, we check that the AC Definition has been primed.
Creation of AC Instance:
Using postman, instance the AC definition using the following template:
To verify this, we check that the AC Instance has been created and is in state UNDEPLOYED.
Creation and deploy of policies and policyTypes:
The AC Instance deploy state is changed from UNDEPLOYED to DEPLOYED using postman:
{
"deployOrder": "DEPLOY"
}
This state change will trigger the creation of policies and policyTypes using the policy-api and the deployment of the policies specified in the ToscaServiceTemplate. To verify this we will check, using policy-api endpoints, that the onap.policies.native.apex.ac.element policy, which is specified in the service template, has been created.
And we will check that the apex onap.policies.native.apex.ac.element policy has been deployed to the defaultGroup. We check this using pap:
Undeployment and deletion of policies and policyTypes:
The ACM STATE is changed from DEPLOYED to UNDEPLOYED using postman:
{
"deployOrder": "UNDEPLOY"
}
This state change will trigger the undeployment of the onap.policies.native.apex.ac.element policy which was deployed previously and the deletion of the previously created policies and policyTypes. To verify this we do a PdpGroup Query as before and check that the onap.policies.native.apex.ac.element policy has been undeployed and removed from the defaultGroup:
As before, we can check that the Test Policy policyType is not found this time and likewise for the onap.policies.native.apex.ac.element policy: