Page Comparison

Table of Contents

Instructions for MANUAL installation Airship+Tungsten Fabric using the Regional Controller and the TF Blueprint

...

Both nodes must be available by ssh. 

You can create your environment manually or This document describes detailed manual installation procedure. As an option you can use Automatic deployment with ansible.  to get the same environment as used for CICD validation.

Overview

Akraino Regional Controller  is necessary part of release 2 deployment procedure. It's Akraino approved blueprint which is common for all of release 2 blueprints and which is using for Edge Site, Blueprint and POD deployment.

...

More information about Regional Controller:

• The Regional Controller's Object Model (it helps you to figure out what is Blueprint, Edge Site, POD, Workflow)
• How to write Blueprints and Workflows
• How to load objects into the Regional Controller
• Frequently Asked Questions

...

You can use  any machine or VM dedicated for this purpose. See instructions on how to start the regional controller.  

https://wikilf-akraino.akrainoatlassian.orgnet/wiki/display/AK/Starting+the+Regional+Controller

After you have a working Regional Controller you have to login on it and follow the steps bellow

...

Update all the environment variables 

Mandatory variables:

• RC_HOST - Regional Controller IP
• NODE - IP address of the node where airhip-in-a-bottle with TungtenFabric would be deployed
• BASE_URL (URL to download ssh key and deploy.sh script)

...

source setup-env.sh
cat objects.yaml.env | envsubst > objects.yaml
cat TF_blueprint.yaml.env | envsubst > TF_blueprint.yaml 

As the result you would get correct yaml files objects.yaml and TF_blueprint.yaml with correct data

ubuntu@ip-172-31-37-160:/opt/akraino-tf$ cat objects.yaml
hardware:
  AWS_instance:
    uuid: 5367a004-71d4-11e9-8bda-0017f00dbff7
    description: AWS Ubuntu Xenial for the TF Blueprint
    yaml:
      todo: AWS instance with >=8 VCPU and >=32GB RAM

edgesites:
  TF_Edgesite:
    description: The demo singlenode TF cluster
    nodes: [ node1 ]
    regions: [ 00000000-0000-0000-0000-000000000000 ]

nodes:
  node1:
    hardware: AWS_instance
    yaml:
      oob_ip: 52.47.109.251

ubuntu@ip-172-31-37-160:/opt/akraino-tf$ cat TF_blueprint.yaml
blueprint: 1.0.0
name: TF Edge Cloud
version: 1.0.0
description: This Blueprint defines an instance of the TF Edge Cloud 
yaml:
  # Required hardware profiles (can match on either UUID or name)
  # Note: UUIDs would likely require a global registry of HW profiles.
  hardware_profile:
    or:
      - { uuid: 5367a004-71d4-11e9-8bda-0017f00dbff7 }
  workflow:
    # Workflow that is invoked when the POD is created
    create:
      url: 'http://172.31.37.160:8000/deploy.sh'
      components:
        # SSH key for remote installation
        - 'http://172.31.37.160:8000/ssh_key.pem'
      input_schema:
        rc_host: { type: string }
        ssh_user: {type: string }
        node: {type: string }
        repo_url: {type: string }
        repo_branch: {type: string }

Clone the api-server repository

...

(optional)

(If you are working on Regional Controller this repo is should be already presented in /opt/api-server/scripts)

This provides the CLI tools used to interact with the Regional Controller. Add the scripts from this repository to your PATH:

...

cat POD.yaml.env | envsubst > POD.yaml

As the result you get correct POD.yaml

name: My_TF_Edge_Cloud_POD
description: Put a description of the POD here.
blueprint: 76c27993-1cc3-471d-8d32-45f1c7c7a753
edgesite: 52783249-45e2-4e34-831d-c46ff5170ae5
yaml:
  rc_host: 35.181.44.122
  node: 52.47.109.251
  ssh_user: ubuntu
  repo_url: https://github.com/progmaticlab/treasuremap.git
  repo_branch: master

Please check that file POD.yaml contains correct data.

Create the POD

Create the POD using

...

where $PODID is the UUID of the POD. This will show all the messages logged by the workflow, as well as the current status of the workflow. The status will be WORKFLOW while the workflow is running, and will change to ACTIVE if the workflow completes successfully, or FAILED, if the workflow fails.

Uninstall

Uninstall Regional Controller

As we using one-time AWS instances they can be just removed with AWS console or other tools which were used for creating (ansible, terraform, etc).

In other cases following comands can be used for manual cleanup procedure.

Deleting POD from Regional Controller

rc_cli -H $RC_HOST -u $RC_USER -p $RC_PW pod delete $PODID

Deleting Blueprint from Regional Controller

rc_cli -H $RC_HOST -u $RC_USER -p $RC_PW blueprint delete $BPID

Uninstall Regional Controller itself

sudo docker stop $(docker ps -aq)
sudo docker rm $(docker ps -aq)
sudo docker rmi $(docker images -q)
sudo rm -rf /opt/api-server/
sudo rm -rf /opt/akraino-tf/

Uninstall Airship

Airship-in-a-bottle doesn't have any tools for installation. Moreover according the documentation it's not recommended to use one virtual instance twice after fail. It's better to remove the failed instance and create a new one for reinstalling.

So the best way to uninstall airship-in-a-bottle it's removing Airship+Tungsten Fabric host via AWC console.