Gitiles
Code Review Sign In
gerrit.nordix.org / onap / aai / model-loader / f2d42f2d0cf695ad317662e5f7d4272efbea56b7
commitf2d42f2d0cf695ad317662e5f7d4272efbea56b7[log] [tgz]
authorMohammed Naser <mnaser@vexxhost.com>Mon Apr 16 13:40:08 2018 -0700
committerJames Forsyth <jf2512@att.com>Wed May 23 20:33:06 2018 +0000
treec3518932ca4cf3d855534ee306e99c67261dac6b
parent543c8a0fea00872c003b8ebe685dfd7ab85ac3b0 [diff]
Pin jnr-unixsocket for Docker plugin to 0.13

The current verison of jnr-unixsocket is 0.12 from the Docker
plugin however in order to be able the images for ARM64, it needs
the newer version which has support for it.

There is work upstream to switch the version to 0.18[0] however
it seems that this is still pending, so we can pin it for now until
that merges.

[0]: https://github.com/spotify/docker-maven-plugin/pull/379

Change-Id: Id863e5fccd4e89f7113b58f8c6b5be760a8797b7
Issue-ID: CIMAN-154
Signed-off-by: Mohammed Naser <mnaser@vexxhost.com>
1 file changed
tree: c3518932ca4cf3d855534ee306e99c67261dac6b
  1. src/
  2. .gitignore
  3. .gitreview
  4. INFO.yaml
  5. License.txt
  6. pom.xml
  7. Readme.md
  8. version.properties
Readme.md

Introduction

The A&AI Model Loader Service is an application that facilitates distribution and ingestion of new service and resource models from SDC to A&AI.

Features

The Model Loader:

  • registers with the SDC to receive notification events
  • polls the UEB/DMaap cluster for notification events
  • downloads artifacts from SDC upon receipt of a distribution event
  • pushes distribution components to A&AI

Compiling Model Loader

Model Loader can be compiled by running mvn clean install A Model Loader docker image can be created by running docker build -t onap/model-loader target

Running Model Loader

Deploying The MicroService

Push the Docker image to your Docker repository. Pull this down to the host machine.

Create the following directories on the host machine:

./logs
./opt/app/model-loader/appconfig
./opt/app/model-loader/appconfig/auth

You will be mounting these as data volumes when you start the Docker container. For examples of the files required in these directories, see the aai/test/config repository (https://gerrit.onap.org/r/#/admin/projects/aai/test-config)

Populate these directories as follows:

Contents of /opt/app/model-loader/appconfig

The following file must be present in this directory on the host machine:

model-loader.properties
# Always false. TLS Auth currently not supported ml.distribution.ACTIVE_SERVER_TLS_AUTH=false

# Address/port of the SDC
ml.distribution.ASDC_ADDRESS=<SDC-Hostname>:8443

# DMaaP consumer group.  
ml.distribution.CONSUMER_GROUP=aai-ml-group

# DMaaP consumer ID
ml.distribution.CONSUMER_ID=aai-ml

# SDC Environment Name.  This must match the environment name configured on the SDC
ml.distribution.ENVIRONMENT_NAME=<Environment Name>

# Currently not used
ml.distribution.KEYSTORE_PASSWORD=

# Currently not used
ml.distribution.KEYSTORE_FILE=

# Obfuscated password to connect to the SDC.  To obtain this value, use the following Jetty library to 
# obfuscate the cleartext password:  http://www.eclipse.org/jetty/documentation/9.4.x/configuring-security-secure-passwords.html
ml.distribution.PASSWORD=OBF:<password>

# How often (in seconds) to poll the DMaaP cluster for new model events
ml.distribution.POLLING_INTERVAL=<integer>

# Timeout value (in seconds) when polling DMaaP for new model events
ml.distribution.POLLING_TIMEOUT=<integer>

# Username to use when connecting to the SDC
ml.distribution.USER=<username>

# Artifact type we want to download from the SDC (the values below will typically suffice)
ml.distribution.ARTIFACT_TYPES=MMODEL_QUERY_SPEC,TOSCA_CSAR

# List of message bus addresses on which to listen for distribution events
ml.distribution.MSG_BUS_ADDRESSES=<host1>,<host2>

# URL of the A&AI
ml.aai.BASE_URL=https://<AAI-Hostname>:8443

# A&AI endpoint to post models
ml.aai.MODEL_URL=/aai/v*/service-design-and-creation/models/model/

# A&AI endpoint to post named queries
ml.aai.NAMED_QUERY_URL=/aai/v*/service-design-and-creation/named-queries/named-query/

# A&AI endpoint to post vnf images
ml.aai.VNF_IMAGE_URL=/aai/v*/service-design-and-creation/vnf-images

# Name of certificate to use in connecting to the A&AI
ml.aai.KEYSTORE_FILE=aai-os-cert.p12

# Obfuscated keystore password to connect to the A&AI.  This is only required if using 2-way SSL (not basic auth).
# To obtain this value, use the following Jetty library to obfuscate the cleartext password:
# http://www.eclipse.org/jetty/documentation/9.4.x/configuring-security-secure-passwords.html
ml.aai.KEYSTORE_PASSWORD=OBF:<password>

# Name of user to use when connecting to the A&AI.  This is only required if using basic auth (not 2-way SSL).
ml.aai.AUTH_USER=<username>

# Obfuscated password to connect to the A&AI.  This is only required if using basic auth (not 2-way SSL).
# To obtain this value, use the following Jetty library to obfuscate the cleartext password:
# http://www.eclipse.org/jetty/documentation/9.4.x/configuring-security-secure-passwords.html
ml.aai.AUTH_PASSWORD=OBF:<password>
Contents of the /opt/app/model-loader/app-config/auth Directory

The following files must be present in this directory on the host machine:

aai-os-cert.p12

The certificate used to connected to the A&AI

Start the service:

You can now start the Docker container for the Search Data Service, in the following manner:

docker run -d \
	-e CONFIG_HOME=/opt/app/model-loader/config/ \
    -v /logs:/logs \
    -v /opt/app/model-loader/appconfig:/opt/app/model-loader/config \
    --name model-loader \
    {{your docker repo}}/model-loader

Where,

{{your docker repo}} = The Docker repository you have published your image to.