Install a distributed scalable environment

 

This document describes how to install an Abiquo distributed scalable environment using the Abiquo OVAs.

 

Introduction

To install any Abiquo environment you should have the following required skills and understand the basics of Abiquo.

Required skills

Some essential system administration skills required for managing the cloud platform are as follows:

  • CentOS/RedHat Linux expertise for maintaining the Abiquo servers, which are Linux servers

  • Computer and network security expertise to configure and maintain the cloud platform

  • Advanced hypervisor knowledge including VM templates

  • Networking including design, switch configuration, VLAN and DHCP and/or SDN configuration

  • Storage including hypervisor datastores and NFS

If you are testing Abiquo, you may not need all of these skills right away, but they are required to run a production environment.


Distributed scalable install profile

There are three main ways to install Abiquo:

  1. Monolithic server, a small environment, also for labs and testing

  2. Distributed scalable, a medium environment with separate components, suitable for production

  3. Distributed HA cluster, a large, high performance environment

This document describes how to install a Distributed scalable environment. 

A Distributed scalable environment means that you have one separate server for each of the server components, and you can later scale some of the components up as clusters by adding more servers.

For more details about the Abiquo install profiles, see Abiquo components.


Diagram of distributed scalable environment

The following diagram shows a distributed scalable environment to give you an idea of how it will look when you have installed it.

Distributed scalable environment
Distributed scalable environment

You can note the IP addresses of your components and other parameters for you installation on the Distributed environment worksheet. For an example of what the parameters could look like, see Distributed environment worksheet example.


Server components

Abiquo has the following key server components.

  1. Abiquo Server with the cloud orchestrator, API, and UI

  2. Datanode services with MySQL database, RabbitMQ, Redis, and other services

  3. Remote Services to manage the cloud resources, with one set per Abiquo DC.

  4. V2V/BPM services to manage VM disk conversions with one per Abiquo DC.

  5. Monitoring optional server to Manage VM and platform metrics data

You will install one of each of these servers in your Distributed scalable environment. For more details, see Abiquo components.


Pre-install

Before you install the servers, you will need to prepare an NFS folder and your networks


NFS folder

The platform will store the catalogue of VM templates in an NFS folder. So make sure to prepare a folder with enough space to hold all your templates and a fast network connection.

The platform will automatically mount the NFS folder on remote services and hypervisors (using the IP address of this server and the folder name that you enter during the install).

When you deploy a VM, the platform copies the template from the NFS to the hypervisor datastore.
The platform can also convert templates from one format to another (V2V).

For more details, see VM repository folder.

Within each datacenter, on the V2V appliance and Remote Services appliance and hypervisors, Abiquo must mount and write to the same VM repository folder


Networks

You will need to put the Abiquo servers and components on the same network, which we call the Management network.

Then the VM networks are a group, which we call the Service network. You can manage the service network with standard networking (DHCP, or guest setup) or using SDN (NSX-T). For these networks you will need the following.

  1. A range of network tags for your VMs to use in private networks.
    Configure this in your network infrastructure and then enter the tags for each group of hypervisors when you create your racks in Abiquo. For each private network, Abiquo will assign a tag from the range.

  2. Network tags of other networks in your infrastructure that users can work with:

    1. Public networks: shared by all enterprises, users can purchase IPs

    2. External networks: any kind of IP address, for one tenant only.

You must also connect your hypervisors to the Service network… :-)

To assign IP addresses to VMs, Abiquo has a default DHCP configuration, which you can use in an isolated test environment. Abiquo also offers dnsmasq. 

For more details, see:


Prepare your hypervisors

Abiquo supports VMware vCenter with access at the cluster or host level. Abiquo supports any kind of datastore that your hypervisor supports. Your hypervisors will need to mount the NFS folder. Abiquo gives users remote access to their VMs via WebMKS, for which we usually install a separate proxy.

For the supported features and configuration details, see VMware.


Install OVAs and configure

We have created the install instructions by stages, so you can install the servers and perform a basic configuration, and then verify each stage with some steps in the Abiquo UI.

  1. Deploy distributed scalable server and datanode

  2. Deploy distributed scalable remote services

  3. Deploy distributed scalable monitoring server

After you configure the OVAs, you can configure many more aspects of the platform. For more details, see Configuration.


Next steps using Abiquo

After you complete and verify your installation, here are some additional links for next steps using your Abiquo platform.

 

 

 

Copyright © 2006-2024, Abiquo Holdings SL. All rights reserved