clearwater.md

Clearwater

Introduction

Clearwater is IMS in the Cloud. See more about this solution !

The Clearwater architecture is designed for deployment on cloud environments. It sourced on the best practice IT software architecture (Load-balancing state-less, easy clustering etc..). See more about the software architecture of Clearwater !

Deployment

To deploy the vIMS Clearwater you must have previously installed the Cloudify orchestrator on OpenStack (see this docs).

Upload clearwater blueprint

Log into the host where you installed the Cloudify CLI and enter in the virtual environment with source command

Download blueprint :

cd ~/cloudify/cloudify-manager/
mkdir blueprints
cd blueprints
git clone -b master https://github.com/Orange-OpenSource/opnfv-cloudify-clearwater.git

Upload blueprint on the orchestrator :

cd opnfv-cloudify-clearwater
cfy blueprints upload -b clearwater -p openstack-blueprint.yaml

Create and launch deployment

Before launch the deployment, it must be created in cloudify.

For that, must be specified the deployment inputs parameters. A template file already exists, you can copy it and complete it to fit with your openstack cloud platform :

cp inputs/openstack.yaml.template inputs/inputs.yaml
vi inputs/inputs.yaml

Bellow an example of input.yaml file configurations:

 image_id: 'ubuntu_14.04'
 flavor_id: 'm1.small'
 agent_user: ubuntu
 key_pair_name: cloudify_ims_kp
 private_key_path: '/etc/cloudify/cloudify_ims.pem'
 external_network_name: ''
 public_domain: clearwater.opnfv
 release: repo122
 bono_cluster_size: 1
 sprout_cluster_size: 1
 vellum_cluster_size: 1
 dime_cluster_size: 1
 homer_cluster_size: 1

Once the completed input file, we must create the deployment on orchestrator :

cfy deployments create -b clearwater -d clearwater-test --inputs inputs/inputs.yaml

Then, you have to launch clearwater deployment :

cfy executions start -w install -d clearwater-test

During the deployment many logs appears on console :

2017-06-27T14:41:03 CFY <clearwater-57> [sprout_host_c68cc.install] Sending task 'diamond_agent.tasks.install'
2017-06-27T14:41:04 CFY <clearwater-57> [sprout_host_c68cc.install] Task started 'diamond_agent.tasks.install'
2017-06-27T14:41:04 CFY <clearwater-57> [bind_296de.configure] Task succeeded 'script_runner.tasks.run'

Once the deployment successfully finish, you can create new numbers on ellis website and then register a sip phone (softphone..) You can help you with this clearwater documentation page (by default the secret_key is "secret").

But to register your softphone, you must use dns of the deployment with his public ip. Or declare the bonos publics IP on your own DNS server. Only bono ip because the private domain fact load-balancing between clearwater VMs.

Test

### Introduction To test your deployment you can use the clearwater-live-test. This test creates numbers on ellis and then made different calls to validate deployment. The result of this test appears on console during the test.

 Basic Call - Mainline (UDP) - (6505550395, 6505550898) Passed
 Basic Call - Tel URIs (TCP) - (6505550449, 6505550796) Passed
 Basic Call - Tel URIs (UDP) - (6505550493, 6505550059) Passed
 Basic Call - Unknown number (TCP) - (6505550405, 6505550862) Passed

Using OPNFV/Functest container

It's the same test but the installation of all dependancies was already done in this container. In addition, this test provide a json file of all test result.

You can run Functest container in your cloudify cli VM. To do that, you must install docker into this VM :

curl -sSL https://get.docker.com/ | sh

After that, you can download OPNFV/Functest container image :

docker pull opnfv/functest-vnf

Then you can run the container :

docker run --dns=<BIND_PUBLIC_IP> -it opnfv/functest-vnf /bin/bash

Next you can launch the signaling testing of your deployment :

cd /src/vims-test
rake test[<YOUR_PUBLIC_DOMAIN_NAME>] SIGNUP_CODE=secret

Scaling

Manually

With built-in workflow on cloudify you can manually scale your clearwater deployment.

Before scale your deployment, create input file with this parameters :

vi scale.yaml

node_id: sprout
delta: 1 
scale_compute: true

This parameters will create one sprout VM and add it on sprout cluster.

Then launch scale with this command :

cfy executions start -w scale -d clearwater-test -p scale.yaml

Warning ! : For the moment, the bono scale doesn't work correctly because Cloudify don't support "one_to_one" relationships. See the post on cloudify forum !

Auto-Scale

 groups:
  clearwater_hosts:
    members: [sprout_host]
    policies:
      mem_scale_policy:
        type: cloudify.policies.types.threshold
        properties:
          service: cpu.total.user
          threshold: 85
        triggers:
          scale_trigger:
            type: cloudify.policies.triggers.execute_workflow
            parameters:
              workflow: scale
              workflow_parameters:
                node_id: sprout
                delta: 1

Monitoring

On cloudify 3.2 non-commercial version, the dashboard is no longer integrate. But, if you want to display clearwater monitoring informations. You can install external Grafana server and then connect this dashboard to InfluxDB API on cloudify-management-server.

After, you can use dashboard model to display all clearwater data :

sed 's/clearwater/'<DEPLOYMENT_NAME>'/g' resources/grafana/clearwater-grafana-dashboard.json > clearwater-dashbord.json

Import 'clearwater-dashboard.json' file on Grafana dashboard.

Uninstall

Log into the host where you installed the Cloudify CLI and enter in the virtual environment with source command

To uninstall properly clearwater deployment, execute this command :

cfy executions start -w uninstall -d clearwater-test