Sorry to have to say this, but I can't recall the last time I read anything in LJ that interested me. Business, web programming and networking take up the bulk of your publication, and I'm sure these things are of interest to many, but not to me. Furthermore, many of the software/book reviews are amateurish and almost all of them lack depth. For example, a review of a math program that actually gave an example of how to do something nontrivial, as opposed to “this program knows how to multiply matrices, isn't that neat” would be welcome. Useful tips for the garden-variety UNIX user would be nice, but that's already covered quite well by Linux Gazette. I also wouldn't mind seeing a well-written article that provided some hard criticism of the shortcomings of various Linux distributions and pulled no punches, but of course you don't want to offend your advertisers.
I am an incarcerated subscriber. Until my recent transfer to a pre-release center I worked with computers for ten years in various industries within the Texas prison system. I have worked with Linux off and on over the past five years. Our computers have tended to be older machines handed down after their primary users deemed them obsolete. Thus, they are more prone to hardware failures. I have learned the hard way that backups are essential. I have come to appreciate a backup and restoration process that is as simple as possible—Keep It Short and Simple (KISS).
I read with interest Charles Curley's article “Bare Metal Recovery” (November 2000). And, I have to credit him with devising a sophisticated method of backing up and restoring a system. Good work! However, his method is too sophisticated, and I balk at the thought of having to repeat such a complex process for each machine that I need to back up. The problem is that the restoration disk contains the backup of the base system (“metadata”, as Mr. Curley calls it) and, therefore, a restoration disk must be built for each and every system. (Not to mention the need to resave the base system to the restoration disk every time the configuration is changed.) The restoration disk should only contain a minimal system with the appropriate drivers and utilities and the restoration software. The backup of the base system itself should be contained on the normal backup media. This way a single restoration disk can be used on any number of machines.
If you are using a backup utility that is too large to fit on a floppy, then building a restoration disk on a Zip drive (or Jaz drive or spare IDE hard drive) is appropriate. However, I noticed that the venerable tar utility won the Favorite Backup Utility category in your “2000 Readers' Choice Awards” (November 2000). Hurrah! tar is my favorite, too. Why? Because with very little effort I can restore a Linux system (or an entire department of Linux boxes) from a single set of boot and root floppy disks. You can build your own floppy-based system (re: “Customize Linux from the Bottom—Building Your Own Linux Base System”, November 2000) or you can use any of the many miniature distributions. I prefer to use Slackware's boot and root disks (I may be a little out of date here, not having used their newer disks since version 4.0.). Their “scsinet” boot disk contains common SCSI and NIC drivers. If you need a driver that is absent, you can build a custom kernel from the source (same version, of course) and write it to the boot disk. Their “rescue” root disk contains tar, file system and network utilities. You may need to create a device node for your particular backup device. I appreciate the convenience of Slackware's boot and root disks. I have never had a need for anything else, and I have had many occasions to restore a downed system or migrate a system to another machine. tar is definitely the backup/restoration tool du jour and a floppy-based system is the easiest means of restoring one or many systems. KISS!
I read your October 2000 “Comparison of Backup Products” by Charles Curley with great interest. The article was very surprising, especially since our product (Microlite BackupEDGE) was not included in the test.
I say surprising because we've been in the Intel UNIX/Linux backup marketplace since 1987. Also, we were the first commercial vendor to introduce a disaster recovery component for Linux products, back in April of 1999 at Comdex Spring.
In short, how did you miss us?
—D. Thomas Podnar, President Microlite Corporationtom@microlite.com
The only reason I did not mention your product was simply because I needed to limit the review in size and scope to keep it down to a manageable size. As you know, there are a lot of good products on the market, and plenty of gnuware and freeware products as well. Had I taken the time to cover all of them, I'd still be at it, and would take up an entire issue of the magazine if I ever got it done. I do hope that my comments on the products I reviewed were sufficiently general that the reader can learn something about evaluating backup software in general.
|Where's That Pesky Hidden Word?||Aug 28, 2015|
|A Project to Guarantee Better Security for Open-Source Projects||Aug 27, 2015|
|Concerning Containers' Connections: on Docker Networking||Aug 26, 2015|
|My Network Go-Bag||Aug 24, 2015|
|Doing Astronomy with Python||Aug 19, 2015|
|Build a “Virtual SuperComputer” with Process Virtualization||Aug 18, 2015|
- A Project to Guarantee Better Security for Open-Source Projects
- Concerning Containers' Connections: on Docker Networking
- Problems with Ubuntu's Software Center and How Canonical Plans to Fix Them
- My Network Go-Bag
- Firefox Security Exploit Targets Linux Users and Web Developers
- Doing Astronomy with Python
- Build a “Virtual SuperComputer” with Process Virtualization
- Where's That Pesky Hidden Word?
- diff -u: What's New in Kernel Development
- Three More Lessons