Build your own cloud with antMan
Workers
Building a private cloud is time consuming and complex. Some use OpenStack, but many OpenStack projects fail because companies underestimate its complexity. antMan does not rely on OpenStack – or any of the other major open source applications for private clouds – but brews its own potion. The tool claims that you just need to install antMan on one or more systems. The rest of the setup is then easy, and the process is largely automated.
Ants Everywhere
Before I get started, a brief clarification of the terms used in the context of antMan will be useful. For the most part, they are contrived and have something to do with ants. The likelihood of knowing the technical terms used here is pretty low if you have never used antMan before. antMan is not, strictly speaking, the name of the Linux distribution on which the product is based. Rather, it is the entire product and the software for managing the physical nodes, which is the core of the product, so to speak. As described here, a Linux distribution named edgeLinux [1] is specially adapted to the needs of this setup. The antHill is the vendor's central registration page, where each newly installed antMan instance is registered with a separate product key.
antMan also distinguishes between antsles and antlets . In terms of content, the two have absolutely nothing to do with each other. An antlet is a virtual instance. The product supports legacy full virtualization with KVM and Qemu and uses the LXC container format propagated by Canonical (but not very well established in the market). In contrast, an antsle is a hardware appliance that the vendor offers bundled with antMan. They come in versions named One, Two, and Nano and are aimed at different user groups.
I won't deal with antsles in this article, but it doesn't hurt to have heard the term. In any case, if you already have an antMan cloud, you will find antlets a fairly simple way to extend it without the need to hunt for hardware.
Getting Ready to Install
The example in this article assumes a physical server with an eight-core CPU, 128GB of RAM, and 4TB of hard disk storage space (RAID 1). For simplicity's sake, it also assumes that direct access to the computer is possible with a keyboard and monitor, which is an option if the corresponding device can be controlled remotely over the Intelligent Platform Management Interface (IPMI), Dell Remote Access Controller (DRAC), Integrated Lights-Out (iLO), or IBM Rational Software Architect (RSA), although it is independent of the operating system. The server needs at least one network interface card with a generous helping of bandwidth and ideally a redundant connection that will be used later as a bridge to implement the network connections for all of the hosted antlets.
The first step is to download the ISO file that installs both edgeLinux and antMan. You can only retrieve the download link from the manufacturer, and because each has a dynamic ID encoded into it, I can't specify that link. Instead, you need to register on the provider's website [1] and supply your email address to receive the link. The antMan package is about 3.1GB, and you should follow the normal steps to put the ISO file on a DVD or USB stick. Most servers today ship from the factory without an optical drive, so USB is the more likely option, and the one I use in this article.
Although the subject of ISOs on USB is not new, problems frequently occur when you try to boot from a USB stick with an ISO file in place, not least because USB sticks behave differently as boot media (e.g., different from CD-ROMs). Moreover, whether or not the Unified Extensible Firmware Interface (UEFI) is enabled on the target system plays a role. In all cases, the recommendation is to use the free Etcher [2] tool to install the ISO file on your USB stick. Pay attention to the obligatory warning that Etcher will delete all the content of the USB stick before writing the ISO file. In the vast majority of cases, however, Etcher leaves behind a USB stick that servers recognize as a bootable device.
Installing edgeLinux
After booting the target system from the USB stick, it becomes your first antMan node. Although edgeLinux is a special development by antMan for its own virtualization solution, you can tell you have CentOS under the hood – even if it is not the latest version. If you have already installed Red Hat Enterprise Linux (RHEL) or CentOS, the process is unlikely to faze you. Even if you are not experienced in RHEL and CentOS, the installation should be pretty easy.
The makers of antMan and EdgeLinux explicitly point out that you need to accept all of the suggestions made by the system regarding partitioning (Figure 1). Furthermore, if you want the system with edgeLinux and antMan to have a static IP address on the local network, you will definitely want to complete the installation before saving it in the antMan configuration, instead of during the installation of the operating system.
If the edgeLinux partitioner does not give you a suggestion for partitioning the system disk, proceed as follows:
- Create a 1024MB partition for
/boot
and a 20GB partition for/
. Leave the remaining space unpartitioned; antMan will take care of partitioning it later. Note that this example assumes the system is equipped with a RAID controller that implements the required RAID 1 in the hardware. Although egdeLinux does support software RAID 1, you will need to configure it yourself when partitioning the disk in the course of the setup. - Make sure you define a password for the root user during installation of the system packages. Remember the password you used, because edgeLinux does not create an additional user with
sudo
rights during the install. If you forget the password, you have no alternative but to start the process from scratch. - This example assumes that the newly installed system has a connection to the Internet after the install.
- After rebooting the system, log in as root and type the password you set earlier. Now wait a few seconds; edgeLinux is configured to start the built-in antMan installer automatically and adjust some settings the first time you log in as root after the installation. Among other things, the program does a system update and guides you through the process of setting up the storage pool for ZFS.
- After the update installation, you simply enter the number of the data carrier whose free space you want to use. If your system has only one (virtual, RAID-1-based) disk, the selection is no big deal. The display on the screen tells you that Ansible will then start and change various system settings. At the end of the process edgeLinux wants to reboot again, so respond yes to complete the setup.
Buy this article as PDF
(incl. VAT)
Buy ADMIN Magazine
Subscribe to our ADMIN Newsletters
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Most Popular
Support Our Work
ADMIN content is made possible with support from readers like you. Please consider contributing when you've found an article to be beneficial.