Slashdot Mirror
VMware Announces UVAC Winners
muff1253 writes to tell us VMware yesterday announced the winners of the Ultimate Virtual Appliance Challenge (UVAC). The contest, which started at the end of February, was designed to test teams on their ability to create a "pre-built, pre-configured, and ready-to-run" application that could be packaged with operating systems in virtual machines.
More info available here.
Information wants a fueled airplane waiting at the hangar and no one gets hurt.
It seems like the top three winners are working in the right direction. I setup a virtual machine at home (albeit using Virtual PC) after Symantec kept quarantining all of the fun tools that I wanted to work with. Virtual machines provide a great environment for setting up network tools that might otherwise not get along with applications and services running on a production server.
First of all, I had no idea what the article was about from the summary. Once I clicked through however I became even more perplexed, for a different reason.
The idea behind the contest is that you build an application bundle that can be run "out of the box" inside of vmware, with no configuration or installation.
So the question is, if you are going to target your application to a virtual machine, why use vmware? Why wouldn't you use java or python, for example?
====
Crudely Drawn Games
One point to it that I could imagine is that two years from now, this post will still be here, but TFA might have moved or disappeared entirely.
Insert self-referential sig here.
I dont know whats sadder, the fact that theres a UVAC acronym, or the fact that I was able to discern what it means simply from the inclusion of vmware in the rss feed.
It looks like all the winners are just some application that already works just fine, in a VM.
That's great and all, but wouldnt it work EXACTLY the same if you did an "install with defaults" on your normal system?
Just saying, you might save 500MB, or even 900MB of download in some cases. One is only 3MB, wow!
- Adam L. Beberg - The Cosm Project - http://www.mithral.com/
Our company uses LEAF in a number of our customers' firewalls. Is your GUI code available for use with "real" LEAF/Shorewall configurations?
We've used Webconf in a *very* limited number of sites. As a rule, our philosophy is that any kind of remote access to a firewall is more risk than we want. However, I would be *very* interested in seeing how your scripts work, and what you have to have open on the LEAF box. If it's less risky than running an HTTP server, it would be something we would consider.
Of course, it would have been nice to see it in something like, say, Python instead of .NET, but beggars can't be choosers! :)
Linux IT Consulting and Domino Development in Michigan
Kennieth et al,
First off, kudos for a very interesting entry & your 3rd place win.. I'm sure it was very well deserved.
I was wondering about the challenges of designing a hybrid OS packet filtering system due to the differences in the IP stack implementation between these 2 OS's. I'm not overly familiar with the ins & outs of the XP IP stack, but with older win32 systems, the differences between the MS stack & the Linux/BSD stacks were significant.
I understand that you basically created a win32 front-end for a *NIX based FW (ideally getting the best of both worlds). Did you run into anything major with this cross stack implementation? Or, is everything based on what the stack of the packet filtering OS (*NIX) & the standards that it adheres to. How does traffic that would be considered "benign" with a *NIX stack, be interpreted by the MS stack?
Or, am I just completely out in left field. If so... Flame on.
Actually we are not manipulating anything at all. Just taking a linux box with shorewall like you would if it was stand alone setup.
So, in short, everything is based on what the stack of the packet filtering OS (*NIX) and the standards that it adheres to!
To answer you question about any problems, we are still looking for feedback from the community as far as bugs and what-not go. Also looking for developers. It's been released as Open Source and can be downloaded either from VMware.com or http://sievefirewall.sourceforge.com./
Kennieth Goodwin
kenny@skyfinet.com
Actually I think it's a Good Thing to always have somebody copy/paste the article into the discussion, so that it becomes part of the thread's permanent archive.
If you go back and read Slashdot stories from more than a year or two ago (always amusing, I strongly recommend it), most of the links to articles are dead. The only threads where you can really read TFA are the ones where somebody pasted it in as a comment.
You do have a point though, it doesn't really deserve a +5 moderation; as long as the person puts "ARTICLE TEXT" in their subject line (which is also a good thing to do!) it's easy enough to find in the the thread if you want to read it, even if it's down at +1 or +2. The only reason to mod it up would be if somebody posted it AC and you wanted to make sure it was readable to people who browse at +1.
So in general, it's definitely karma-whorish, but on the other hand it's also rather useful...so who cares if people get some free points?
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
http://www.vmware.com/vmtn/appliances/challenge/wi nners.html
http://www.getsieve.com/
As a coworker of the winning team, the main reason for doing an appliance version (apart from participating in this contest) was packaging. We actually do have a "native" windows port of the code (using python, pygtk etc.), and it's about 25 MB zipped when containing all the dependancies, of which there is really way too many to ask a random user to install so it all has to be packaged into the same thing, really.
The vmware image is about 72 MB bzip2-compressed which includes a stripped-down Ubuntu, X11 etc. And it runs on Windows, any random Linux distro that might have an old pygtk/cairo/whatnot that doesn't work with our code, OS X (with OS X vmware) out of the box. Nice even if you do lose some performance and run into issues inherent to virtualization (accurate timestamps and promiscuous mode inside the virtual machine are tricky and do have limitations!).
We mostly run and develop it natively ourselves (on FC5 and OS X), yet we run into "AAARGH! How do I get
a new enough Y for OS X to run this" discussions every week or so.
I want to be able to run a browser in a VM , and play WMV videos. Does one of these do that, I know Mplayer may be illegal in the US, so is Freespire the only one that has it? Or is there another virtual appliance that has a browser and can play WMV?
At first glance I thought it read "UNIVAC winners".
*Phew*
"We are all geniuses when we dream"
- E.M. Cioran
The VMware web site often gives the impression that the company employs a lot of people who have no understanding of computers. The announcement has no links to the winners! The web pages don't display well in Firefox. There are numerous other flaws.
If I didn't already know that VMware is a reputable company, I would never buy anything from a company with such a clueless web site. Obviously someone at VMware thinks that non-technical people have something valuable to contribute to a technical company, even though they cannot understand what they are doing.
Winner: HowNetWorks
Second Place: Trellis NAS Bridge Appliance.
Third Place: Sieve Firewall
Yes, I know, awkward, systems aren't written that way. But we can change the rules, can't we?
Do not mock my vision of impractical footwear
15 servers to 3...now all you need is a couple bits of bad RAM or a slowly melting I/O cable run a bit too close to the super-hot CPU and you are going to really find out how fun this "new" technology (from 1960) is. Worse, virtualizing legacy systems to keep them running is a formula for stagnation and the death of your business. Just ask AT&T (formerly SBC formerly PacBell) which had to run its S.O.R.D. (service order) application which was written in IBM 7074 auto-coder and had to stay running on S/370s in "emulation" mode for decades thereafter because they had lost the original code and some of the patches --in the end, there was a roomfull of 300lb DMV-like women whose job was to keep all of the knowledge of how it worked. They ran like that for decades and decades until finally starting to rewrite everything from scratch. So now it is so much easier to run your 1990s Java or ACT! application in a "virtual" environment --hey, why not keep running it that way until 2016 or 2026? Well, nobody ever "plans" to do that, but the path of least resistance has a way of making its point.
Bottom Line: Young programmers are more productive than older programmers, so the industry tends to be like "Logan's Run". Unfortunately, this means that the industry is forever caught in a "loop", reinventing everything over and over and over again. You know what I say? Fuck the industry. Do you know how much freaking money there is to be made just cleaning up the messes created by 20-something Dreamweaver jockies who started cutting and pasting PHP code until they came up with something that sort of worked?