Transfer from Server > Server > Client without download to server

Question

Here's my case: I have a file on a server (e.g www.server1.com/file.rar). I want send this file to my client. But, the www.server1.com accept connections from only one IP. I have a lot of clients, so I have a lot of IP's.

The solution could be transfer the file from www.server1.com to www.server2.com (which has one IP address only) and provide the file: www.server2.com/file.rar But... In this case I need transfer the file from server1 to server2, and that's (bandwidth and space).

There's any alternatives for that? I was thinking about something like:

header("Content-Disposition: attachment; filename=" . urlencode($file));    
header("Content-Type: application/force-download");
header("Content-Type: application/octet-stream");
header("Content-Type: application/download");
header("Content-Description: File Transfer");

file_get_contents("http://server1.com/file.rar");

But didn't work as expected.

Thanks in advance

Edit: I can't reply my own question cause I'm new user... Well, i want say thanks to Everyone for the replys, they send me to the solution. Which is: file_get_contents didn't work because of server flush, so I used fopen and fread with ob_flush and flush

header("Content-Disposition: attachment; filename=" . urlencode($arr[1])); //$arr[1]
header("Content-Type: application/force-download");
header("Content-Type: application/octet-stream");
header("Content-Type: application/download");header("Content-Description: File Transfer");
ob_flush();
flush();
$fp = fopen($url, "r");//$url variable with file url
while (!feof($fp)) {
    echo fread($fp, 65536);
    ob_flush();
    flush();
}
fclose($fp);

Thanks!

I don't understand. You want to transfer the file from server1 to server2 to the client without actually transferring it to server2? — Gabe, May 22 '11 at 05:31
Well, maybe... I just want a way to save bandwidth and disk space on server2. — StiveKnx, May 22 '11 at 05:38
While you'll save disk space by not downloading it and memory by flushing inside your loop, you'll not save any bandwidth and you'll be paying for server1 -> server2 and server2->client. For data to go from server1 to client without going through that route, you'd have to have direct access from the client. — Lucas, May 23 '11 at 11:33

score 3 · Answer 1 · answered May 22 '11 at 05:54

here some example code. ive not tested it tho but think it would work. using cURL and mod rewrite hides true location of file.

<?php 
function downloader($url){
    $file_name = basename($url);
    $generated_file = geturl($url, $url);
    file_put_contents($file_name,$generated_file);
    $size=strlen($generated_file);
    if($size==0){die("<b>Error:</b>Location to download not found!");}
    header('Content-type: application/force-download');
    header('Content-Disposition: attachment; filename=' . $file_name);
    header('Content-Transfer-Encoding: binary');
    header('Expires: Sat, 26 Jul 1997 05:00:00 GMT');
    header('Cache-Control: must-revalidate, post-check=0, pre-check=0');
    header('Pragma: public');
    header('Content-Length: ' . $size);
    readfile($file_name);
    unlink($file_name);
}

function geturl($url, $referer) {
    $headers[] = 'Accept: image/gif, image/x-bitmap, image/jpeg, image/pjpeg';
    $headers[] = 'Connection: Keep-Alive';
    $headers[] = 'Content-type: application/x-www-form-urlencoded;charset=UTF-8';
    $user_agent = 'ONLY_MY_HOST';
    $process = curl_init($url);
    curl_setopt($process, CURLOPT_HTTPHEADER, $headers);
    curl_setopt($process, CURLOPT_HEADER, 0);
    curl_setopt($process, CURLOPT_USERAGENT, $useragent);
    curl_setopt($process, CURLOPT_REFERER, $referer);
    curl_setopt($process, CURLOPT_TIMEOUT, 30);
    curl_setopt($process, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($process, CURLOPT_FOLLOWLOCATION, 1);
    $return = curl_exec($process);
    curl_close($process);
    return $return;
}


downloader('www.server2.com/file.rar','http://www.server1.com');

?>
-CONTENT SERVER PART 
<?php 
/*
.htaccess
RewriteEngine On
RewriteCond %{SCRIPT_FILENAME} !-d
RewriteRule ^([^\.]+)$ serve.php?file=$1 [NC,L]
*/

//serve.php
if($_SERVER['HTTP_REFERER']=='http://www.server1.com' && $_SERVER['HTTP_USER_AGENT'] =='ONLY_MY_HOST' && $_SERVER['REMOTE_ADDRESS']=='123.123.123.123'){
    readfile('./file_store/'.$_REQUEST['file']);
    die();
}
header('Location: http://www.server1.com');
?>

I think readfile act in the same way as file_get_contents, my problem (maybe) was with flush, which I solved and added in my post. — StiveKnx, May 22 '11 at 06:04
k sorry missed you sayin that while i was writing. glad you sorted now — Lawrence Cherone, May 22 '11 at 06:06

score 2 · Accepted Answer · edited May 23 '17 at 09:59

2

You can use readfile (see http://php.net/manual/en/function.readfile.php). Note that readfile will only work if your server is properly configured (as discussed at PHP readfile() of external URL).

Alternatively you can use echo file_get_contents("http://server1.com/file.rar");.

edited May 23 '17 at 09:59

Community

1
1

answered May 22 '11 at 05:35

Femi

64,273
8
118
148

file_get_contents won't work if readfile doesn't, and since it's a much worse solution, I don't recommend your last line. – Emil Vikström May 22 '11 at 05:37
True, it is a worse solution: its in there for completeness. `readfile` is far superior. – Femi May 22 '11 at 05:48

score 1 · Answer 3 · answered May 22 '11 at 05:33

Use readfile instead of file_get_contents. Readfile will send the file to the client in the same rate it can download it from server1. that way you don't need to keep the file in memory (as with file_get_contents) and your client can start her download before you have fetched the entire file.

Notice that this will download the file to your server but not save it to disk, but instead send it out to your client. Thus the bandwidth usage will be there, but not the disk usage. There's no way for you to solve this without downloading the file from server1 to server2.

If you cache the file on server2's disk, you'll save bandwidth usage on the expense of disk.

score 0 · Answer 4 · answered May 22 '11 at 05:41

0

You can either save bandwidth to server2 by caching the file to server2's disk, or you can save space on server2's disk by always downloading the file from server1 whenever the client requests it. There is no way to do both simultaneously.

answered May 22 '11 at 05:41

Gabe

84,912
12
139
238

The file will be requested only one time for one client, will be rare see another client calling the same file. (server1 has a loooot of files) Clients are going ask for different files, each one for hes needs, so, store on the disk is not useful. Thanks for the reply. – StiveKnx May 22 '11 at 05:51

Transfer from Server > Server > Client without download to server

4 Answers4

Linked