Re: Amazon S3 and Windows 2003 - Is is possible with Cygwin

["Tom Metro-12" <user=71@nabble.blu.org>]

 James Kramer wrote: 
> I was trying to think of an easy way to back up my file systems at 
> work that run off a Win 2003 server.  I would like to use the Amazon 
> S3 online storage. 
> ... 
> Can I use Cygwin to run the Linux version of the Amazon S3 script and 
> back up my windows directories. 

Perhaps, but before going that route I'd explore these options first: 

Take a look at the S3 "Solutions Catalog"[1] and find an S3 client that 
is designed to run on Windows. JungleDisk[2], for example, is a popular 
choice that runs on multiple platforms, but isn't free. 

If you can't find something suitable that works natively on Windows, I'd 
next take a look at some of the solutions written in multi-platform 
languages, such as Perl, Java, or Ruby. These will run on Windows 
without the Cygwin compatibility environment. This blog post[3] lists a 
bunch of them. 

1. 
http://solutions.amazonwebservices.com/connect/kbcategory.jspa?categoryID=66
2. http://jungledisk.com/
3. http://jeremy.zawodny.com/blog/archives/007641.html


> It there any special parameters that I need to use to differentiate 
> between Linux and Unix files--I know that Windows stores files 
> slightly differently than *nix. 

The line endings on text files are different, but any backup system that 
properly preserves the content of files, regardless of where it is 
stored, should transparently store Windows files. What characters are 
legal in file names also differ, but generally UNIX is less restrictive 
than Windows, so using a tool design for UNIX won't impose any problems. 

There are some Windows-specific options for rsync, but if you're going 
to be backing up to S3, you won't be using rsync. (For a bit about why, 
see [4][5].) At best you'll be using a tool that approximates the 
benefits of rsync. (It looks like their are some people using Amazon's 
Elastic Compute Cloud to build true rsync services layered on top of S3, 
but I don't know if any of these services are currently available or 
cost effective.) 

4. http://guillaume.filion.org/blog/archives/2007/02/rsync_algorithm_s3.php
5. http://www.google.com/search?q=s3+rsync


> For that matter, if I were to back up my windows partition... 

Usually when someone speaks of backing up a partition, they're referring 
to doing a low-level image of the partition, including partition tables 
and the file system structure. None of the S3 tools are designed to do 
this, that I'm aware of. They're all designed to work with individual 
files. While you could using a partition imaging tool to create a file, 
and then one of the tools above to transfer it to S3, it would be very 
bandwidth inefficient unless the tool was specifically designed to split 
up the source file into multiple files (chunks) on the S3 side. 

  -Tom 

-- 
Tom Metro 
Venture Logic, Newton, MA, USA 
"Enterprise solutions through open source." 
Professional Profile: http://tmetro.venturelogic.com/

-- 
This message has been scanned for viruses and 
dangerous content by MailScanner, and is 
believed to be clean. 

_______________________________________________ 
Discuss mailing list 
[hidden email] 
http://lists.blu.org/mailman/listinfo/discuss