Cron, file_get_contents(), URL file-access

I've encountered a problem while setting up a cron job.
The script runs fine if I open it in a browser.
When cron executes it it returns the following error messages through email:

Warning: file_get_contents(): URL file-access is disabled in the server configuration in .......... on line 180
Warning: file_get_contents(.............): failed to open stream: no suitable wrapper could be found in ................ on line 180

Code (markup):

Line 180 is the following:

$xmlstring=file_get_contents($feed);

Code (markup):

Where $feed is the url of an XML-file.

Would really appreciate any help on why this suddenly doesn't work.

Thank you.

 

If you work with xml files why don't you use simplexml_load_file() function?

 

The xml-feeds can be quite large which makes simplexml_load_file() to exceed my allowed memory size.

Instead I use:
$xmlstring=file_get_contents($feed);
$xml=new SimpleXMLElement($xmlstring);
Which works fine (well, until now ).

 

Anyone got any clue?

 

I tried another approach, with curl instead:

$ch=curl_init();
$timeout=5;
curl_setopt($ch,CURLOPT_URL,$feed);
curl_setopt($ch,CURLOPT_RETURNTRANSFER,1);
curl_setopt($ch,CURLOPT_CONNECTTIMEOUT,$timeout);
$file_contents=curl_exec($ch);
curl_close($ch);
$xml=new SimpleXMLElement($file_contents);

Code (markup):

This returns another error when running cron (works fine in browser):

Fatal error: Cannot instantiate non-existent class:  simplexmlelement

Code (markup):

Seems weird that it does recognize the class in a browser but not when executed by cron.

 

Try this command in the cron job:

wget -q --spider http://example.com

 

You must set allow_url_fopen to on in your .htaccess or curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);

 

You or your host is running a different php version and settings for CLI (command line/cronjobs) than the Apache (http browser).

 

You need to ticket your web hosting company, allow_url_fopen is a php feature that they can disable, the CLI may use a different PHP configuration and not have the related extension loaded, especially true if you are using something like PHP Selector in the web stack.

 

Problem:
Your server configuration doesn't allow function file_get_contents to read anything other than local files, and your script is trying to read an http url.

Assumption:
It is your script and you coded it so you know how to modify it.

Solution:
a) Use same curl code that you posted in your 2nd reply and that stores curl result in variable $file_contents

  $ch=curl_init();
  $timeout=5;
  curl_setopt($ch,CURLOPT_URL,$feed);
  curl_setopt($ch,CURLOPT_RETURNTRANSFER,1);
  curl_setopt($ch,CURLOPT_CONNECTTIMEOUT,$timeout);
  $file_contents=curl_exec($ch);
  curl_close($ch);

Code (markup):

b) Download attached file called xml2array.php, and use in your script to create php array out of the read XML, this way

      require_once 'xml2array.php';
      $array = xml2array($file_contents);
      //check parsed XML
      print_r($array); // << use this array

Code (markup):

I hope it helps.. check attachment.

Stay well....