Wikipedia:Reference desk/Archives/Computing/2010 June 18

= June 18 =

Virtualization with Microsoft Windows Server 2008 R2 Enterprise Edition
I have a question in regard of creating and hosting Virtual Machines on the Microsoft Windows Server 2008 R2 Enterprise Edition operating-system. The Microsoft website states that Microsoft Windows Server 2008 R2 Enterprise Edition only supports a maximum of four virtual machines. Is this limit of four virtual machines an overall limit, or is it the limit of the number of Windows-based virtual machines which can be hosted by a server with the Microsoft Windows Server 2008 R2 Enterprise Edition operating-system? In other words; can more than four non-Windows based virtual machines run in a single physical Microsoft Windows Server 2008 R2 Enterprise Edition operating-system environment?
 * It doesn't say that. What it says is "Windows Server 2008 R2 Enterprise licenses include the use right to run up to four additional virtual instances of Windows Server with one server that is licensed with Windows Server 2008 R2 Enterprise." That means, I gather, that the purchase price of this product includes four Windows OS licenses intended for use in VMs. It has nothing to do with the number of VMs the product can support. -- BenRG (talk) 00:26, 18 June 2010 (UTC)

Management of Microsoft Hyper-V Server 2008 R2
Is there any software other than Microsoft System Center Virtual Machine Manager 2008 R2 which can be used to manage Virtual Machines running on a server operating on Microsoft Hyper-V Server 2008 R2? In addition, does anyone know what precisely is the licensing model for Microsoft System Center? The Microsoft System Center homepage can be found here, but I have not been able to figure out the licensing system up to now. It's not anything like the regular model of a server operating-system and Client Access Licenses.

Objective & Comparison
My objective is to create and host dozens of Linux-based virtual machines on a single physical machine. Which of the two hypervisor operating-systems; Microsoft Windows Server 2008 R2 Enterprise Edition or Microsoft Hyper-V Server 2008 R2; would be more appropriate for this objective? My key considerations are as follows.


 * 1) Compatibility with both Ubuntu Linux and Debian Linux
 * 2) Ease of Management of Virtual Machines
 * 3) Simplicity of Acquisition - Licensing Model
 * 4) Economy - which one is cheaper and/or more cost-effective? Although Microsoft Hyper-V Server 2008 R2 is free to download, it requires Microsoft System Center to manage the virtual machines hosted on it. And I have to date been unable to decipher the licensing model of Microsoft System Center.

Thanks in advance, Rocketshiporion (talk) 00:12, 18 June 2010 (UTC)


 * I don't think anyone here has any experience with these products. However, I don't think System Center Virtual Machine Manager is required to use Hyper-V Server; is there a reason you think it is? Also, is there a reason you're considering only Microsoft products? Xen and VMware's products are other options. -- BenRG (talk) 07:48, 18 June 2010 (UTC)


 * I believe that Microsoft System Center Virtual Machine Manager is required to manage Microsoft Hyper-V Server as it is stated on this page under the question "Will System Center Virtual Machine Manager (SCVMM) support Microsoft Hyper-V Server? When will it be available?" that "System Center Virtual Machine Manager 2008 R2 is required to manage Microsoft Hyper-V Server 2008 R2. It is available today. An eval version of SCVMM is currently available or download from http://www.microsoft.com/scvmm". I have not yet considered VMware, as I am mostly familiar with Windows and a bit familiar with Linux. Is there any open-source Linux hypervisor which can be used instead of Microsoft Hyper-V Server; i.e. something with similar features? Rocketshiporion (talk) 08:07, 18 June 2010 (UTC)
 * I think the quoted text means that 2008 R2 is the earliest version of SCVMM that will work with Hyper-V Server, not that SCVMM as such is required to use Hyper-V Server. Xen is an open source (GPL) hypervisor with corporate backing from Citrix Systems. I've never used it (or any of these products) but I think it's mature and well regarded. I think its support for Windows guests is rather poor compared to MS and VMware's products, but it supports Linux rather well. -- BenRG (talk) 19:33, 18 June 2010 (UTC)


 * If you're looking for more information, consider posting your questions here or here. Right now, you seem to be talking to an OSS guy.--Best Dog Ever (talk) 01:43, 19 June 2010 (UTC)


 * Thanks. The Microsoft Technet Forum turns out to be quite useful. Rocketshiporion (talk) 06:48, 19 June 2010 (UTC)

Can a whole disk be copied?
I have been struggling to repair a laptop with a totally screwed Windows XP operating system, probably caused by a head crash some time ago. I am particularly keen to keep the other programs which are installed on the same disk, programs for which I do not have replacement installation disks. After successfully recovering (most of) the registry, and using chkdsk to try to patch up the hard disk, the machine no longer boots (stopping with a black screen with just the mouse pointer on it). I have tried using a XP installation disk to repair the existing Windows installation, hoping to repair the broken files that chkdsk identified. However, about halfway through the process it suddenly stopped, saying the hard drive is corrupt and it cannot continue. I was wondering if it is possible to copy the whole disk to folder on another machine (obviously taking the disk out and using an external USB housing to do this), reformat the faulty disk so it correctly marks the damaged sectors as unusable, then copy the original contents back again? I know that will still leave me with broken files, but the idea is to fix the disk corruption that prevented the repair installation from completing. Are there any things likely to trip me up later (for example: what happens if 'C:\System Volume Information' ends up with different attributes from the original)? Astronaut (talk) 03:09, 18 June 2010 (UTC)


 * It's worth trying. Norton Ghost can do a sector copy of your drive to (a) another drive, or (b) a file.  You could then reformat the bad drive and copy the stuff from (a) or (b) back to it, preferably sector-by-sector.  (List of disk cloning software lists other software that can do some or all of this.)  However, I strongly caution you not to use that drive anymore.  If it actually has bad/damaged sectors and this isn't just a software problem, then more bad sectors will appear.  I'd advise you to replace the drive.  Comet Tuttle (talk) 03:27, 18 June 2010 (UTC)


 * If you know *nix (linux/unix) at all, it's very easy. You could get a Knoppix live disk and then run dd if=/dev/sda | gzip -9 -c >> DRIVEIMAGE.img.gz. This will direct copy the disk (in this case SDA, adjust for your purposes), compress it with gzip, and then pipe it out to a file. Adjust the output appropriately so it doesn't overwrite something. This creates a byte-for-byte copy of the original image. If you have to write it back, it's easy to reverse that command. Shadowjams (talk) 08:17, 18 June 2010 (UTC)


 * Just in my own experience, GParted is exceptionally reliable and easy to use for this sort of thing. It does not require you to type in any obscure Linux commands to work. I've used it many times (as a non-Unix user) to make clones from one drive to another. Whether it will work exactly right depends on more than that, but that's a pretty easy thing to do. --Mr.98 (talk) 20:17, 18 June 2010 (UTC)


 * If the disk is physically damaged, ddrescue may be of interest. --NorwegianBluetalk 20:27, 18 June 2010 (UTC)

Is is possible to navigate the Rolling Stone website at all?
When I go to read music reviews or articles and then I read the comment section, sometimes the comments are a little long to display in its entirety, so I have to click on the comment to finish reading it, and instead I get redirected to Rolling Stone's main page. What the hell? 24.189.90.68 (talk) 05:58, 18 June 2010 (UTC)


 * Please provide an explicit example for us to contemplate. -- SGBailey (talk) 12:50, 18 June 2010 (UTC)


 * This is common when reading archived pages on a web site, including our own Ref Desk archives. What happens is that the pages to which the link pointed have moved and it can't find the new location of the (now archived) page, so instead of giving you a general "page not found" error, they redirect you to their home page.  Yes, it is annoying and there are ways to archive without breaking links.  One way is to put everything in it's final archive position right away, initially providing links from the main page, then sever those links from the main page once it has been "archived", but leave all the internal links in place.  Another option is to have a bot go through and search for any broken links created when a page is moved, and modify those links to point to the new location.  Finally, anyone moving a page could be forced to leave a redirect behind, so any links to the old page can still find the new page. StuRat (talk) 15:23, 18 June 2010 (UTC)

Video_game_console vs. laptops
Why do video game consoles exist at all? At the beginning, a couple of decades ago, when computers were expensive, they certainly had a purpose. But, nowadays, why would someone not buy a laptop instead of a video game console? Laptops can do everything and more, can't they?--Mr.K. (talk) 09:29, 18 June 2010 (UTC)
 * You're entering into a long debate my friend. Short version of the argument is that consoles can force developers to write programs a certain way and can also optimize graphical functions. Hence why lots of parallel ps3s can do good distributed projects. That said, most major studios also like to have cross-platform compatibility, including PC. The second piece is the economics. Many consoles are priced below cost on the expectation of selling games for them. Like buying a printer and then buying ink. Or a cellphone, and buying service. Shadowjams (talk) 09:40, 18 June 2010 (UTC)
 * Also, because consoles are entirely built-for-purpose, and their purpose is not as broad as that of a PC, it is generally considered that they have less problems due to conflicts or system inadequacies. That doesn't however make it less likely that an issue is present with the hardware itself, although you would expect it to be put under a lot of scrutiny. Findstr (talk) 11:41, 18 June 2010 (UTC)
 * There have been cases where I've bought a game and then was later disappointed that it wouldn't run on my computer. Hardware and software differences mean that games will not run the same on every system, and sometimes won't run at all. Except for a few rare exceptions, games designed for a particular game console will always run on that console. The hardware and software is all standardized. If I buy a game for my console, I don't have to worry about the requirements. I know it will run because I have the console. I don't have to compare the game's requirements to my machine's specs, or mess with settings to find the ideal configuration. I just stick the disc in and start playing. And the PC is freed up to do something useful while I'm wasting my time playing games. Reach Out to the Truth 20:35, 18 June 2010 (UTC)


 * A large part of it is simply marketing.  A lot of promotion is put into game consoles (Both to consumers and to developers), but there doesn't exist any single company who could do the same for PC gaming.  (Microsoft would be the best bet, but they've already got a console.) APL (talk) 21:59, 18 June 2010 (UTC)

Citing PDFs
Not sure if this is entirely the best place to put it, but USAir Flight 405, an article I'm working on, was just turned down for WP:GA because the number of inline sources was too low. I have based a good deal of the article on the NTSB report on the accident, and most of the quotes in the article come from this PDF. However, I don't want to have the article littered with [1] s, because it would look messy and improper. Is there a way to have inline citations link directly to a certain page in the PDF? I know the URL doesn't change depending on what page you are on, so I was wondering if there was any other way. Thanks in advance,  WackyWace  talk to me, people 11:52, 18 June 2010 (UTC)
 * You can add #page=pagenumber to the end of the URL, though this doesn't always work in every browser. Recent Internet Explorer and Firefox releases should support it, though. NTSB report, page 12. Note that this will use the actual PDF page numbers, which aren't always the same as the numbers printed on the page you're displaying. so the previous link will show pages 3 and 4 of the document, as they are displayed on the 12th page of the PDF file. -- 109.193.27.65 (talk) 12:14, 18 June 2010 (UTC)


 * Or you could just cite the page in the citation ("Report on XYZ, page 6."), without the hotlink. That's an easy way to do it that doesn't depend on PDF readers behaving correctly. Perhaps a better way, on the whole. --Mr.98 (talk) 17:02, 18 June 2010 (UTC)


 * If this led to the article being quick-failed, I would ask the reviewer how he or she would prefer the article to be referenced - and if there was any chance of him or her reconsidering their decision.  (talk) 15:56, 18 June 2010 (UTC)


 * What they want, I gather, is not just a million "general" citations, but citations that say, "this statement comes from page 6." It's not an unreasonable thing to ask for, though it does create a lot of work for someone else, of course. --Mr.98 (talk) 17:02, 18 June 2010 (UTC)


 * True that. A recent discussion on WT:GAN seemed to conclude that page numbers should normally be given. However, there's many ways of doing that. For example the "shortened" style as seen in WP:CITESHORT. I encouraged the author to contact the reviewer to ensure that he or she didn't do it in a manner that would result in another failed review.  (talk) 17:47, 18 June 2010 (UTC)

HTML web page creation using ASP
I can create an HTML formatted web page using ASP code and save it to the server logical path. I can also create an HTML formatted web page manually using a text editor on the client and save it to the server using the URL path. However, what I want to do is to save the HTML formatted web page I can create using ASP not to the logical path on the server but to the URL path. How do I do this? 71.100.0.224 (talk) 14:02, 18 June 2010 (UTC)
 * When you save it to the URL path, would I be right in thinking that you're doing this with FTP? --Phil Holmes (talk) 17:02, 18 June 2010 (UTC)
 * When I save it from the client I can use DOS FTP or FrontPage FTP of FrontPage HTTP since the server is FrontPage and ASP enabled. The problem is in saving a file generated by an ASP program on the server to the URL path I manually access client side by saving using FTP or HTTP post. 71.100.0.224 (talk) 21:21, 18 June 2010 (UTC)


 * If you can post your page to your website, you could send it via the HTTP POST method. The basic syntax is shown here .  Alternatively you could try using the inbuilt FTP exe as shown here: .  If you can't do that, there are a number of other suggestions that you could try by googling "asp ftp". --Phil Holmes (talk) 10:30, 19 June 2010 (UTC)

Firefox RAM
With 10 tabs open (normal wikipedia pages, not games or fancy javascript processes or anything else) Firefox 3.6.3 is using 300mb of RAM after 5 mins of use. Is this normal? How could I reduce this without using just one tab as that basically defeats the point tabs. Are older versions of Firefox (versions 1 or 2) less resource hungry? 82.43.90.93 (talk) 16:00, 18 June 2010 (UTC)


 * On my Mac, I opened up 10 Firefox tabs of CNN.com (a pretty resource-heavy page), and got about 200MB of RAM. I did the same in Safari and got the same result. When it is 10 pages of Google.com, it's only 80MB or so (50MB on Safari). So a lot of that is likely dependent on the page content, how many extra "helpers" the browser has to load to display it. (In terms of "raw size," CNN and Google are pretty similar: 200KB of content for CNN, 103KB for Google. But in terms of page complexity—amount of elements to render, calls made to other scripts, Flash ads and etc.—the CNN page is much more complex.) I suspect this is probably pretty standard but I'd be interested to hear how other browsers perform. I think it is probably unavoidable though that modern browsers rendering modern webpages are very resource-intensive. --Mr.98 (talk) 18:05, 18 June 2010 (UTC)


 * Most resource usage complaints for Firefox can be traced down to its addons - try disabling all the addons and do the same test again, see if there's a significant difference. If there is you can re-enable them one by one until the problem come back. But if you aren't really running out of RAM, reducing Firefox's RAM usage probably won't do much at all. --antilivedT 08:11, 19 June 2010 (UTC)

Greasemonkey scripts
Is it possible for a single greasemonkey script to take different actions depending on different url you're viewing? For example, I have the following two separate scripts; // ==UserScript== // @include      http://en.wikipedia.org/* // ==/UserScript== document.title = 'YOU ARE ON EN [DOT] WIKIEDPIA [DOT] ORG';

// ==UserScript== // @include      http://google.com* // ==/UserScript== document.title = 'YOU ARE ON GOOGLE [DOT] COM';

Is there a way I can combine the function of these two separate scripts into one script? Installing potentially thousands of separate scripts for every website I might want to alter the title of seems like a waste of time if one single easily updateable script can do it. Thanks for your help 82.43.90.93 (talk) 16:15, 18 June 2010 (UTC)
 * Something like this? (My Regex is bit rusty, apologies for any inadvertent error)


 * --59.95.103.172 (talk) 16:32, 18 June 2010 (UTC)


 * The script doesn't seem to be working :( Thank you anyway for trying 82.43.90.93 (talk) 19:55, 18 June 2010 (UTC)
 * Works for me after a slight modification to the Regex :)

It is also possible to "extract" the name of the website from the URL; that way it will work on all websites and without an if...else if ladder. --59.95.99.64 (talk) 20:17, 18 June 2010 (UTC)
 * AWESOME! Thank you so much! 82.43.90.93 (talk) 20:44, 18 June 2010 (UTC)

Plustek OpticFilm Scanner 7500i and 7600i
Can anyone find me some real concrete differences between these two models, apart from the price tag?

The plustek website and review site do not indicate this and I have found it pretty frustrating after quite some research.

Anyone have anything concrete?

86.140.210.188 (talk) 18:14, 18 June 2010 (UTC)


 * The Plustek does have a feature "comparison" page. Primary differences seem to be the lamp (7600i has a white LED, 7500i uses a cold cathode), different preview speeds (7500i is faster), and scanning speeds (7500i is faster except at 7200 dpi, for some reason). --Mr.98 (talk) 20:12, 18 June 2010 (UTC)

I wasn't signed in but I had no history
If I look at my history with CTRL-H I can only see today and yesterday. Earlier this week, there was no yesterday since I had not been on the computer in several days, yet the first two Wikipedia pages I went to showed I was signed in. The third said I wasn't. Perhaps it had been 30 days, but there shouldn't have still been pages in my history if I can't see my history beyond the previous day--right? Vchimpanzee ·  talk  ·  contributions  · 21:20, 18 June 2010 (UTC)


 * I often see "unusual" Wikipedia claims that I am logged in after days away from the computer. I believe it's just because I've read the page before and it's been stored locally in the browser cache, and the copy of the page that is in the cache says at the top "Comet Tuttle - my talk - my preferences"...etc.  I think the third page you visited in your example above was a new page you didn't have in your browser cache.  If this bothers you, you can clear your browser cache.  Comet Tuttle (talk) 22:30, 18 June 2010 (UTC)


 * Well, like I said, when I do CTRL-H there is no history beyond "yesterday". So how could it still be in my browser cache? That's what I'm trying to answer. I used to clear my history daily but then I found out there wasn't any left after I had been gone several days. Or so I thought. Vchimpanzee ·  talk  ·  contributions  · 15:31, 19 June 2010 (UTC)


 * The history and the browser cache are separate things. You can see this in Internet Explorer 8 by going to Tools -> Internet Options, and on the General tab, in the "Browsing history" section, click "Settings".  You'll notice that there's a "Temporary Internet Files" section, where you can tell it how much disk space to use to store copies of web pages you visit; and, separately, the "History" section lets you tell it how many days to keep a history of the pages you visited.  If the history is set to, say, 3 days, and you come back 5 days later, the history will show as empty; but everything in the Temporary Internet Files cache will still be there.  Comet Tuttle (talk) 17:25, 19 June 2010 (UTC)


 * So I should still be deleting every day or at least frequently to keep it from getting cluttered. Vchimpanzee ·  talk  ·  contributions  · 18:35, 21 June 2010 (UTC)

When is a gigabyte not a gigabyte?
When a company advertises certain storage devices, they count the bytes in a base of 10 to yield a greater number of gigabytes than if they were counted in a base of two and this leads to confusion when ignorant end-users install these devices and discover that they appear to offer lower capacity than advertised. Does this apply to all data storage devices, or just HDDs? What about SSDs? Just because it appears to be standard in the industry, does that mean it must remain an acceptable practise? Could it not be deemed misleading by the UK Office of Fair Trading, requiring all sellers to describe their goods in binary quantities of bytes? --78.150.225.204 (talk) 22:25, 18 June 2010 (UTC)


 * This "issue" has occurred in the US and there have been class action lawsuits about the allegedly "missing" storage, and the issue is behind the recent use of the odious term gibibytes and its symbol "GiB". Comet Tuttle (talk) 22:32, 18 June 2010 (UTC)


 * Some space from the advertised size is always lost by system files and file tables etc 82.43.90.93 (talk) 22:34, 18 June 2010 (UTC)


 * I think that ram is usually still advertised in the traditional base two units, (Partially because it needs to be packaged that way, for technical reasons.) But most storage cards are sold advertised in the SI base ten units.
 * I would assume that the reason it's not considered fraudulent is because of the IEEE 1541-2002 standard which attempts to define the SI base ten units as correct, and give the traditional base two units silly names.
 * Adaptation of this standard has been haphazard outside of storage manufacturers. Many claim that using base ten units for computer storage is unintuitive, while others claim that it was politically motivated by hardware manufacturers. APL (talk) 22:41, 18 June 2010 (UTC)


 * "Giga" almost always means 109. The only widespread exceptions are (a) RAM sizes and (b) sizes reported by commonly used file management software like Windows Explorer. The software could easily be changed to use decimal units instead; in my opinion, this should have happened long ago. -- BenRG (talk) 03:09, 19 June 2010 (UTC)


 * This issue gets worse as memory, disk space, etc. get bigger. Here's a chart of the difference between the base 10 and base 2 units at different scales:

KB 2.4% MB 4.9% GB 7.4% TB 10.0%


 * So, while 2.4% is small enough to ignore, a 10% difference begins to become significant. 68.248.75.49 (talk) 04:53, 19 June 2010 (UTC)II
 * They do it with things as small as pen drives. I know my "2 GB" pen drive is actually about 1.9 gigs, because of this rounding. I think everything should be rounded down to decimal units, instead of up to binary. As software becomes more accessible, so should its workings and terminology. {&#123; Sonia &#124;ping&#124;enlist}&#125; 05:11, 19 June 2010 (UTC)
 * It's not roundng. It's different units.  And the lower number is when it's reported in binary.  So if you have 2 GB/1.9GB then the 2GB is decimal while the 1.9 is binary.  Taemyr (talk) 15:59, 20 June 2010 (UTC)