What you will learn in this tip: As Linux has become essential to many businesses, the need for disaster recovery...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
for Linux-based systems has grown. Learn about the variety of Linux disaster recovery options for your organization in this tip.
The good news about Linux disaster recovery is the number and variety of Linux disaster recovery tools have grown in the last couple of years. Today there are a wide variety of disaster recovery options available.
Writing shell scripts
You can do Linux disaster recovery the old-fashioned way by writing a series of shell scripts. Due to the highly modular nature of Linux and other Unix-like operating systems, this is easier than it sounds and a number of the disaster recovery products available are mostly a collection of scripts. However, writing a bunch of scripts to automate disaster recovery from scratch isn’t really cost-effective.
One wrinkle in Linux disaster recovery systems is the wide range of rescue media you can create to store the basic information. This is possible because the Linux boot system is extremely compact by modern operating system standards. You can put a basic Linux boot system, partition information and other basic image data on a USB drive as well as the more conventional set of CDs. If a Linux server becomes unbootable, you can often restore it simply by plugging in a thumb drive in the USB port. (Which is why USB boot drives have become a status symbol on Linux sysadmin’s keychains everywhere.)
Almost all of the major backup and recovery software packages (like Symantec Corp. Backup Exec, for example) support Linux. In addition, there are a number of disaster recovery options written specifically for Linux. These are both open source and commercial.
There are a number of freeware disaster recovery options available (like rsync), many of them with quite sophisticated features. For example, Relax and Recover, or ReaR, is a Bash script for disaster recovery available from SourceForge and others. The result is a highly modular disaster recovery framework with no dependencies on things like the GUI. And besides several internal and generic methods to integrate with backup software, ReaR includes integration modules with many of the major backup software packages (e.g., CommVault, Hewlett-Packard, IBM Corp, Symantec). The disaster recovery information can be stored via network, DVDs, USB devices, tape, etc.
Commercial disaster recovery options for Linux systems
Commercial products such as SBAdmin from Storix Inc. are also available with a full range of features and support options for Linux disaster recovery. Like its Windows counterparts, SBAdmin also handles conventional backup, image cloning and most other backup and recovery chores. With it, you can provide a bare-metal recovery of Linux systems.
There are also backup appliances from companies like Unitrends that handle both disaster recovery and conventional backup with deduplication in a single box. In addition to backup, these appliances allow you to do bare-metal restores of systems.
Linux disaster recovery can also be done in the cloud. Companies like rackAID offer data recovery as a service using the cloud to back up to multiple remote data centers. With rackAID, you can restore single files, directories, or your entire server. Many of these cloud-oriented companies also offer managed online backups, disaster recovery planning and backup testing.
Many cloud-based and conventional remote disaster recovery products use a snapshot or continuous data protection (CDP) process to create the disaster recovery image and keep it current. This means frequently transmitting small amounts of data which have changed since the last image update was done. Recovery can present more of a problem because of the large amount of data that has to be restored in something like a bare-metal recovery. Here you are limited by the bandwidth of your connection to the cloud and that may mean a long recovery period, although some cloud services will overnight your data on a disk or even a server.
With any disaster recovery software it’s important to understand just which jobs a package will do. This is even more important in the Unix/Linux space. For reasons ranging from operating system design to philosophy, Unix/Linux has a lot of specialized tools that aren’t really intended to do a complete disaster recovery.
For example, the open-source package Clonezilla is highly regarded for producing images of a complete file system. Clonezilla is very useful for jobs like loading a preconfigured image onto a new desktop PC or server, but it is clumsy for taking regular images of a production system for DR purposes. Among other things, you can’t produce an image of a system that’s active, which means you would have to do your cloning at a time when the system was inactive.
When choosing any Linux disaster recovery product it's important to understand the capabilities of the product and how it will work within your environment.
About this author: Rick Cook specializes in writing about issues related to storage and storage management.