You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 21 Next »

Contents

Introduction

This document describes the steps to create a single and multi-edge node. Edge Nodes are deployed at physical locations called Edge Nodes.

Unicycle (Three-Node Cluster) - Supported in the future release

Edge Node Configuration

Rover (Single-Node Cluster)

This configuration file is written as a series of shell key/value assignments.

Sample input file template with all required parameters are provided in the ‘Build or Deploy an Edge Node' section.

Key Name(s)

Notes

Example

SRV_NAME

Server hostname
SRV_NAME=aknode42

SRV_OOB_IP SRV_OOB_USR SRV_OOB_PWD

Out-of-band interface information for the server (iDRAC, ilo, etc.)
SRV_OOB_IP=192.168.41.42
SRV_OOB_USR=root
SRV_OOB_PWD=XXXXXXXX

SRV_MAC

MAC address of the server used during the build
SRV_MAC=3c:fd:fe:b8:02:90

SRV_OEM

Server OEM: Dell or HPE (case sensitive)
SRV_OEM=Dell

SRV_IPXE_INF SRV_BLD_INF SRV_BOOT_DEVICE

Network interface used during the build when ipxe.efi and the OS are booted
SRV_IPXE_INF=net4
SRV_BLD_INF=enp94s0f0
SRV_BOOT_DEVICE=sdg

SRV_BLD_SCRIPT

ipxe script based on the OS version and target kernel. Valid values are script-hwe-16.04.4-amd64.ipxe or script-16.04.4-amd64.ipxe

SRV_BLD_SCRIPT=script-hwe-16.04.4-amd64.ipxe

SRV_BIOS_TEMPLATE SRV_BOOT_TEMPLATE SRV_HTTP_BOOT_DEV

XML template used to set BIOS and RAID configuration.
SRV_BIOS_TEMPLATE=dell_r740_g14_uefi_base.xml.template
SRV_BOOT_TEMPLATE=dell_r740_g14_uefi_httpboot.xml.template
SRV_HTTP_BOOT_DEV=NIC.Slot.7-1-1
SRV_FIRSTBOOT_TEMPLATE

Template for script to run on first boot to setup the server. Current options are:

firstboot.sh.template
firstboot-genesis.sh.template
firstboot-airship-iab.sh.template

SRV_FIRSTBOOT_TEMPLATE=firstboot-airship-iab.sh.template

SRV_VLAN

VLAN to use during the build and final network configuration
SRV_VLAN=41

SRV_MTU SRV_IP SRV_SUBNET SRV_NETMASK SRV_GATEWAY SRV_DNS SRV_DOMAIN SRV_DNSSEARCH SRV_NTP

Basic network information for DHCP configuration and final server network settings
SRV_MTU=9000
SRV_IP=192.168.2.42
SRV_SUBNET=192.168.2.0
SRV_NETMASK=255.255.255.0
SRV_GATEWAY=192.168.2.200
SRV_DNS=192.168.2.85
SRV_DOMAIN=lab.akraino.org
SRV_DNSSEARCH=lab.akraino.org
SRV_NTP=ntp.ubuntu.org

SRV_PWD

Root password for the server being built
SRV_PWD=XXXXXXXX

SRV_BOND SRV_SLAVE1 SRV_SLAVE2

Network bond information
SRV_BOND=bond0
SRV_SLAVE1=enp94s0f0
SRV_SLAVE2=enp94s0f1


Populating PostgreSQL

The Akraino seed code comes with default two sites: MTN1, MTN2 representing two lab sites in Middletown, NJ. This step of connecting to the database and creating edge_site records are only required if the user wishes to deploy on other sites.

To deploy a Rover (Single-Node Cluster) Edge Node, perform the following steps:

  • Check if the Akraino (Docker Containers) packages are stood up.
  • Connect to PostgreSQL database providing the host IP (name).
jdbc:postgresql://<IP-address-of-DB-host>:6432/postgres
user name = admin
password = abc123
use ‘pgAdmin |||’ Postgres client tool or connect to Postgres DB using SQL interface  


  • Execute the following SQL insert, bearing in mind these value substitutions:
    • edge_site_id: Any unique increment value. This is usually 1 but does not have to be.
    • edge_site_name: Human-readable Edge Node name.
    • region_id: Edge Node region number. Use select * from akraino.Region; to determine the appropriate value. Observe the region number associations returned from the query: Use 1 for US East, 2 for US West, and so on.

> insert into akraino.edge_site(edge_site_id, edge_site_name, crt_login_id, crt_dt, upd_login_id, upd_dt, region_id)
  values( 1, 'Atlanta', user, now(), user, now(),1);


Akraino Portal Operations

Login

Visit the portal URL http://REGIONAL_NODE_IP:8080/AECPortalMgmt/ where REGIONAL_NODE_IP is the Portal IP.

Use the following credentials:

  • Username: akadmin
  • Password: akraino

Upon successful login, the Akraino Portal home page will appear.

Build or Deploy an Edge Node

From the Portal home page:

  1. Select one Edge Node.
  2. For the selected Edge Node, select a Blueprint (Rover or Unicycle) from the drop-down menu.
  3. Click on Upload button (in the Blueprint column), this will open a pop-up dialog
  4. Provide the edge site-specific details such as IP address, user, password.
  5. Click on Browse button, select the input file that supplies the input parameters as defined in ‘Edge Node Configuration - Rover (Single-Node Cluster)’ the input file is a property file that stores information in key-value format.  Sample input file used for ‘Rover’ deploy: Rover.txt
  6. Click on Submit. This will save the input file and site details.
  7. When using the Unicycle blueprint, the Build button is enabled.
    1. Select Build to begin the build process.
    2. The Build status is updated to Build Complete, once the process is finished.
    3. Observe that the Deploy button is enabled.
    4. Select Deploy to begin the deploy process.
    5. The Deploy status is updated to Completed once the process is finished.
  8. When using the Rover Blueprint, the Deploy button is enabled.
    1. Click on Deploy to begin the deploy process.
    2. This process takes about an hour.
    3. User can click on Refresh (link) to update the status of the deploy on the portal.  
    4. The deploy status changes from ‘Not started’ to ‘In progress’ to ‘Complete’

Tempest Tests

Prerequisites

Specifics around OpenStack are beyond the scope of this guide.

  • Validate all OpenStack settings as needed.
  • Ensure all necessary OpenStack services are up and running for a particular site.
  • Use appropriate values for the OpenStack username, password, region, and desired timeout.      

Running Tempest

  • Navigate to ETE testing page using the side menu in the portal.
  • Select a deployed site in the Run Tests page.
  • Click on Run Tempest to trigger tempest test
  • Test status gets updated to 'Complete', once the job gets completed.

Installing ONAP

Prerequisites

Specifics around ONAP are beyond the scope of this guide.

  • Verify that a VLAN is created and can be used for OpenStack tenant VM
    • The VLAN ID
    • An IP (or a range of IP) that can be accessed from this VLAN
    • Gateway, DNS, for the IP
  • A public private key pair for accessing ONAP VM
  • If the Edge Node is behind proxy, the proxy URL
  • Obtain the OpenStack keystone admin password

Configuration

This configuration file is written as a YAML dictionary named parameters.

This dictionary will be selectively inserted into the ONAP VM environment file, depending on overall configuration.

Key Name
Notes
Example
public_net_nameThe name assigned to the OpenStack network that will be created and used by the ONAP VM. An existing network may be used by referencing its name here.
public_net_name: public
public_physical_net_provider_nameThe physical provider network name for the public network. CAUTION: If the physical provider network cannot be found, the automated deployment process will terminate with an error.
  # Assume vlan50 is the VLAN created 
  # for the OpenStack Project (tenant)
  public_physical_net_provider_name: vlan50
provider_segmentation_idThe VLAN tag of the physical provider network
  # Assume vlan50
  provider_segmentation_id: 50
public_physical_net_typeThe physical network type. Specify VLAN or flat. NOTE:When flat is in use, provider_segmentation_id will be ignored. 
  public_physical_net_type: vlan
public_subnet_nameThe subnet name. An existing subnet may be used by referencing its name here.
  public_subnet_name: public-subnet
public_subnet_cidrThe subnet’s CIDR. The ONAP VM will be assigned an IP within this subnet. 
  public_subnet_cidr: 192.168.102.0/24
public_subnet_allocation_startThe allocation start of the IP pool. Together with public_subnet_allocation_end it defines the range of IPs assigned to VMs.
  public_subnet_allocation_start: 192.168.102.100
public_subnet_allocation_endThe allocation end of the IP pool. Together with public_subnet_allocation_start it defines the range of IPs assigned to VMs.
  public_subnet_allocation_end: 192.168.102.200
public_subnet_dns_nameserverThe subnet's DNS server
  public_subnet_dns_nameserver: 8.8.8.8
public_subnet_gateway_ipThe subnet's Gateway IP
  public_subnet_gateway_ip: 192.168.102.1
flavor_nameVM flavor for the ONAP VM. The installation creates m1.onap, x1.onap, xx1.onap flavors, all of which are recommended for use.
  flavor_name: x1.onap
onap_vm_public_keyThe public key used to access the ONAP VM. A URL pointing to the key may be used here.
  onap_vm_public_key: PUBLIC_KEY
http_proxyOptional http proxy URL including port. If the network is accessible without a proxy, please leave empty.
  http_proxy: http://proxy.yourdomain.com:8888
https_proxyOptional https proxy URL including the port. If the network is accessible without a proxy, please leave empty.
  https_proxy: https://proxy.yourdomain.com:8888
no_proxyOptional list of domains, IPs, and CIDRs that do not require a proxy, separated by commas. Do not use spaces. If the network is accessible without a proxy, please leave empty.
  no_proxy: 127.0.0.1
keystone_admin_passwordThe keystone admin password. 
  keystone_admin_password: XXXXXXXX
onap_tenant_passwordThe keystone password for the ONAP project (formerly known as a tenant in OpenStack parlance). 
  onap_tenant_password: XXXXXXXX
onap_artifacts_http_repoOptional HTTP repository URL for ONAP artifacts. When deploying a self-contained Edge Node without Internet access, all ONAP artifacts must be downloaded from ONAP Nexus to a HTTP server, using this URL. If HTTP is not being used to download artifacts, please leave empty.
  onap_artifacts_http_repo: http://192.168.102.220/onap

Installation

  1. Navigate to the add-ons page via the sidebar menu.
  2. Select a deployed site on the Install ONAP page.
  3. Choose Install ONAP to begin installation.
  4. The ONAP status changes to Complete once the job is finished.
  5. ONAP VM will be created successfully after the job is finished. It takes several hours until ONAP becomes operational.

Onboarding a Sample VNF

  1. Navigate to the Akraino Sites page.
  2. Select a deployed site. The VNF Onboard button becomes enabled.
  3. Choose VNF Onboard and supply the VNF Type.
  4. Choose Onboard to begin the sample VNF installation.

The Sample VNF creates an Apache Traffic Server cache instance and a Locust load generator client instance in the OpenStack environment for the selected site.  By default, the Sample VNF will create a heat stack name ats-demo.  The heat stack will contain two instances named ats-demo-client and ats-demo-server.  The default configuration of the server instance is to cache the Big Buck Bunny videos from the internet server and then serve the video to the client from its ramdisk. The heat stack will output the client and server URLs.  The client URL will display the Locust load generator dashboard.  The server URL will allow you to play the video in a browser.

OpenStack stack output show ats-demo --all

Field      

Value                                                                                 

client_url

{

"output_value": "http://192.168.48.223:8089/", 
"output_key": "client_url", 
"description": "URL for client"

}

client_ip

{

"output_value": "192.168.48.223",
"output_key": "client_ip", 
"description": "IP of the client"

}

server_url 

{

"output_value": "http://192.168.48.226:8080/bbb_sunflower_1080p_30fps_normal.mp4",
"output_key": "server_url", 
"description": "URL for server"

}

server_ip  

{

"output_value": "192.168.48.226",
"output_key": "server_ip", 
"description": "IP of the server"

}


VNF can also be on-boarded using ONAP. Refer the documentation.


  • No labels