|
 |
|
|
| Author |
Message |
Andrew Guest
|
Posted: Thu Dec 29, 2005 2:56 am Post subject: Information about download accelerator |
|
|
|
We’ve just brought download accelerator, we`ve installed it.... now what? We are unable to put our usual websites that we download with into the add URL box. Are we doing some thing wrong or do the web sites that we use meant to automatically come up?
It is designed to download files from the internet.
When using download accelerator, you have to put in the actual website address with an exe/zip etc extension. (The file you wish to download)
[b:8c3f030877]For example[/b:8c3f030877]
[url]http://dld91.winmx.com/8198751285/winmx353.exe[/url]
You type this in the URL section of the program and click on start download. This will download the file to the specified path. To find out what the URL is. Click on the download link within the website with your right mouse button and select [b:8c3f030877]COPY SHORTCUT[/b:8c3f030877]. Then paste this shortcut into the program Download Accelerator URL section.
|
|
|
|
|
|
|
|
| Back to top |
|
 |
1888 Discuss Guest
|
Posted: Thu Dec 29, 2005 2:56 am |
|
|
|
|
|
| Back to top |
|
 |
Pires Guest
|
Posted: Thu Dec 29, 2005 5:37 am |
|
|
When your web browser tries to download something from a site, it opens up a connection to the site's HTTP server and downloads the file through that connection. This is how things are supposed to work.
One of the swell features of modern HTTP servers is the ability to request certain "ranges", or parts, of a file. The idea behind this is that, if you get disconnected while you're downloading a file, you can resume the download later and not have to start over.
Download accelerators capitalize on the idea of resemble downloads by downloading multiple ranges at the same time, hoping against hope that this will somehow make things go faster. Once in a blue moon, when the site that the file is being downloaded from is artificially capping download speeds, a download accelerator will actually work. It will work by bypassing limits which were probably put in place for a very good reason, but it'll work.
To download multiple ranges at the same time, the download accelerator has to open multiple connections to the HTTP server. This is bad, because HTTP servers can only handle a finite number of simultaneous connections, and so when one person is opening up four or eight or twenty connections, he's filling up spaces that could otherwise be used to allow that many other people to connect to the server. Get enough people doing this, and soon no one will be able to connect to the server.
To prevent this is to limit the number of simultaneous connections to the HTTP server from any given IP address. In My case, a single IP address is allowed to open up three connections simultaneously. This works quite well to protect the HTTP server from the abuse that download accelerators throw at it.
Now enter "Download Accelerator Plus", probably the most popular download accelerator around. When it starts a download it does its thing, flooding the server with connection attempts and trying to download certain pieces of the file. After three connections, My HTTP server sends it a "503 Service Unavailable" message. This is supposed to mean that the client (i.e. DAP) should stop trying to connect to the server for some period of time. And so DAP does stop trying to connect. But what it also does is to close the existing, perfectly legitimate connections to the HTTP server. Why it does this, I'm not entirely sure. But the net effect is that it's more-or-less impossible to download anything using DAP.
At any rate, the solution is to stop using Download Accelerator Plus. Lots of folks evidently use DAP solely because it can resume downloads (the ability to resume downloads is a good, useful feature which is beneficial to both the server and the client), and they can't get that functionality in any useful form from their web browsers. There are lots of other more reasonable, more clue fully designed programs which can resume downloads. Get Right ([url=http://www.getright.com]www.getright.com[/url]), for one. |
|
| Back to top |
|
 |
|
|