session_write_close

bernard dot savonet+ at gmail dot com (2012-03-15 21:46:07)

Beware of session_write_close when calling PHP scripts from an Ajax page: this does not "finalize" the session data writing, but in fact it resets it!

anton dot kril at gmail dot com (2012-03-05 11:40:09)

session_write_close can have no effect with debugger(XDebug) turned on. session_write_close doesn't unlock session for other connections. Noticed on WAMP with php 5.3.8

sss at activators dot info (2011-10-06 01:55:25)

i have been using a mysql database custom session handler (not using cookies) and was having problems getting the session data to be saved consistantly on my database driven website
clients also have to navigate across domains for website management at times so i have some special code to make all this happen relatively seamlessly and to ensure that each person's data is secure.
i added session_write_close() at the very end of the last script that set session data and solved the problem.
i am not sure why, but it seems the calls to write and close were not always being made (i was not smart enough to figure it out)
now that the session_write_close() call is being made, my problems seemed to have disappeared - hopefully for good.
hope this helps someone.

joadha at gmail dot com (2011-03-10 10:23:20)

Remember: objects with private and/or protected data members cannot be serialized properly by PHP. If you ever find that you are missing some objects from your $_SESSION, make sure you are accounting for this shortcoming in your application.

Anonymous (2010-03-25 11:44:57)

You can have interesting fun debugging anything with sleep() in it if you have a session still active. For example, a page that makes an ajax request, where the ajax request polls a server-side event (and may not return immediately).
If the ajax function doesn't do session_write_close(), then your outer page will appear to hang, and opening other pages in new tabs will also stall.

prophp at gmail dot com (2009-11-26 06:21:04)

Beware, if you overwrite the default PHP Session handling and use debugging code inside the write() function, the debugging code is not executed until you run session_write_close().

I tried everything, file logging directly from the write() function, global debugging variable increments, static class properties. The only things written were the session open() and read() calls. My debugging code looks like this:
<?php
$Session = new Session();
...
class Session() {
    public function write($id)
        $sql = "UPDATE ... WHERE id=". mysql_real_escape_string($id);
        self::$debug_Info .= "session_write sql=$sql";
    ...
}

# then at the very end of the script:
# session debugging
session_write_close(); 
error_log($Session->getDebugInfo(), 3, 'logs/sessions.log');
?>

where getDebugInfo simply returns self::$debug_Info. Without session_write_close() the sessions.log would only contain the open() and read() calls.

Maybe intuitive to many, it took days to realize. hope it helps!

web at murray-mint dot co dot uk (2009-09-11 05:36:22)

If you're saving data to a session but finding it's not actually being saved, check and ensure that you're not assigning any arrays with a key containing the pipe character (|). This will prevent the session data from being serialized and saved.

martyrs_cry at hotmail dot com (2009-05-14 13:06:56)

Using a SQL Server database and Windows Server 2003, if you use session_write_close() prior to a header() call, and it still appears to freeze up your session, check to make sure that you are not in the middle of a transaction.
I noticed that by starting a transaction, executing some queries, attempting to write close the session and then redirecting without rolling back or committing that transaction, your session may freeze up. Just like what you may experience if you attempt to redirect with header() without first calling session_write_close().

noel (2008-11-03 21:29:53)

I was having the same problem as many here regarding setting session data just before a header location redirect and having the session data just not be there. I tried everything people here said, and none of their combinations worked. What did finally work for me was to fire off a session_regenerate_id(true) call just prior to the header() and die() calls.
session_regenerate_id(true);
header('location: blah blah');
die();
Without the regenerate id call, the write close did not seem to do anything. session_write_close() doesn't seem to matter at all. It certainly didn't fix anything on its own for me.
This is a rather annoying issue with php sessions that I've never run into before. I store my sessions to /dev/shm (which is RAM) so file IO blocking can't be the problem. Now I'm nervous that some other session data might not be getting updated prior to a header() location change, which is extremely important and common in any web app.

ivan at alexandrov dot biz (2008-09-29 07:14:16)

I had a problem with realizing the restore password form. First a user entered his login or e-mail in the system.

Then the script searched the database, got the session data, and sended link with SID to registered e-mail. The link was configured so, that it restored session data and logged user in the secure interface to the change password form.

Then was displayed a page with the message about sended message.

The problem was that ID was not unique in three pages, the SID sended to e-mail anyone could see in cookie.

I tryed to start new session before generating and after sending link with the code:

<?php ....
session_start();
/*Getting user login and e-mail from database*/
$user_login = "....";
$user_id = "....."

/*CLOSE PREVIOUS SESSION*/
session_unlink();
session_destroy();

/*NOW GENERATING LINK FOR SESSION DATA */
session_start();
$_SESSION = $user_login;
$_SESSION = $user_id;
/*here generating link:*/
$link = "http://host.com/restore=" . SID . "";
mail (....);

/*CLOSE THE SESSION WITH USER DATA*/
session_write_close();

/*AND STARTING A NEW SESSION*/
session_start();
/*THEN LOAD THE 'MESSAGE SENDED' PAGE*/
header("Location: /restore/message_sended/");

?>

The trouble was that SID was the same even after session_unlink() and session_write_close(). The session_start() function just restored the previous session data!!! So the script was not safe. 
Then I added session_regenerate_id() call after each session_start().

<?php ....
session_start();
/*Getting user login and e-mail from database*/
$user_login = "....";
$user_id = "....."

/*CLOSE PREVIOUS SESSION*/
session_unlink();
session_destroy();

/*NOW GENERATING LINK FOR SESSION DATA */
session_start();
session_regenerate_id();//Regenerating SID for sending

$_SESSION = $user_login;
$_SESSION = $user_id;

/*here generating link:*/
$link = "http://host.com/restore=" . SID . "";
mail (....);

/*CLOSE THE SESSION WITH USER DATA*/
session_write_close();

/*AND STARTING ANOTHER NEW SESSION*/
session_start();
session_regenerate_id(); //Regenerating SID
/*THEN LOAD THE 'MESSAGE SENDED' PAGE*/
header("Location: /restore/message_sended/");

?>

 And now it works as needed! The SID sending to user we cannot see in cookies nor before neither after generated link, but the data is saved in session with this id. So only the owner of account can get it!

kent openit se (2008-09-22 02:49:15)

When trying to use exec on Windows 2003 Server together with WAMP you probably will experience that the server stops to answer your requests.
Through calling session_write_close before -every- exec this will solve your problem.
Hope this will help someone!

yacine at wasabihawaii dot com (2008-08-06 02:04:20)

My sessions were screwing up because I had 2 different session IDs going at once:
- 1 PHPSESSID cookie for domain.com
- 1 PHPSESSID cookie for www.domain.com
This fixed it:
//At the beginning of each page...
session_set_cookie_params (1800,"/","domain.com");
session_start();

lee (2008-06-20 07:35:53)

It seems that the session data is written out when there is actual changes to the array.
class Login(){
function logoff(){
session_unset();
@session_destroy();
@session_start();
$_SESSION = array();
session_commit();
}
}
//during test, this doesn't work and the session data remain.
classLoginObject->logoff();
//I fix it like this
class Login(){
function logoff(){
session_unset();
@session_destroy();
@session_start();
$_SESSION = array();
$_SESSION['cleardata'] = 1; <------
unset($_SESSION['cleardata']);<------
session_commit();
}
}
//after the fix, this works and the session is empty.
classLoginObject->logoff();
I was using phpunit(Sebastian Bergmann) to do my test.

burak >>>at (2008-04-25 23:32:21)

I had a similar problem with session data.
I lost the session data randomly, without any pattern. I didn't even use the header(location:... ) function.
Tried:
set "session.use_cookies" to 1 in php.ini
session-write-close()
ob_start() / ob_end_flush()
No matter what I did, it didn't worked.
Finally, when I set the "session.use_only_cookies" to 1, problem is solved. I am no longer losing any sessions.

infinite____ at abv dot bg (2007-11-11 02:10:28)

I had a problem with sessions and header("Location: URL"). The session cookie was created, however every time I call session_start() a new session file was created. Please note, that the solution is in php.ini: session.referer_check is the option. I have changed it to a location x and when I tried to start up session in location y, the session cookies were invalid.

steve at dot-totally dot co dot uk (2007-07-21 12:03:58)

For the session problem when using header("Location:..."), I found session_write_close() not to help me on the my IIS server using PHP in CGI mode. The problem was the PHPSESSID cookie was never being set, so I did it the manual way:
header("Set-Cookie: PHPSESSID=" . session_id() . "; path=/");
Worked for me this way!

(2007-04-26 15:11:47)

I was having trouble with session data being lost / not carrying over to other pages when using header("Location: ..."), and session_write_close() was not helping.
I was writing to $_SESSION by doing:
$_SESSION = $myArray;
I eventually found that doing it this way helped:
foreach($myArray as $k => $v)
{
$_SESSION[$k] = $v;
}
The problem was strange; I've assigned arrays to $_SESSION countless times and never had trouble with it. In fact, I was doing it in many places elsewhere in the same application, and I never ran into any problems.
I spent hours figuring out what was happening. Hope this saves time for somebody out there.

cottoANTISPAM at ori dot org (2007-03-14 11:37:18)

This function is useful for forcing serialization of session data but it can introduce difficult-to-track bugs if it's called more than once per session_start() call.  Since it doesn't have a return value or raise an exception there won't be any indication that the serialization failed and the code will continue normally.  Only when a user visits a page that depends on unsaved session data will there be any indication of the failure.
As far as I can tell this affects both the default and custom session handling functions.

<?
session_start();
$_SESSION['foo'] = 'box';
session_write_close();

if (array_key_exists('visited',$_SESSION) {
    echo "welcome back!\n"; # will never happen
} else {
    echo "I don't know you.\n";
    $_SESSION['visited'] = true;
    session_write_close();
}
?>

tyler at n49 dot com (2007-01-31 17:49:36)

Along the same lines as what cenaculo at netcabo dot pt, bkatz at usefulengineering dot com, and editorial at literati dot ca said about making sure you session_write_close(), don't forget to ob_end_flush() if you're using output buffering.

I've been having some weird hanging issues when I tried to navigate away from a page with content streaming in an Iframe or a separate window.

Instead of:

<?
ob_start();
session_start();
/* Do session stuff here; security; logging; etc. */
session_write_close();
/* NOW write out the requested file. */
header("Content-type: audio/x-mpeg"); /* or whatever type */
header("Content-Disposition: attachment; filename=" . $filename);
header("Content-Length: " . $filesize);
header("Content-Transfer-Encoding: binary\n\n");
header("Pragma: no-cache");
header("Expires: 0");
$file_contents = file_get_contents($filepath);   
print($file_contents);
?>

Do:

<?
ob_start();
session_start();
/* Do session stuff here; security; logging; etc. */
session_write_close();
/* Make sure data is actually flushed out to the browser */
ob_end_flush();
/* NOW write out the requested file. */
header("Content-type: audio/x-mpeg"); /* or whatever type */
header("Content-Disposition: attachment; filename=" . $filename);
header("Content-Length: " . $filesize);
header("Content-Transfer-Encoding: binary\n\n");
header("Pragma: no-cache");
header("Expires: 0");
$file_contents = file_get_contents($filepath);   
print($file_contents);
?>

risaac at deadletter dot com (2006-04-04 12:46:12)

Workaround if session_write_close() still doesn't write sessions fast enough:
I found with one PHP login system that even session_write_close() was not setting the session variables before I transferred pages with a Location: header. So the user would log in, I would create the $_SESSION variables, call session_write_close() and then transfer to the secure page using header(Location:...). The secure page would check for the session vars, not find them, and force the user to log in again. After the second login the session would be found and they could continue.
My workaround was to create the $_SESSION variables with 0 values before writing the initial login page. Then I updated the session vars with the login results and used the header() function to switch to the secure location. Once the session vars have already been created, updated values are assigned quickly. Problem solved. Just be sure the secure page checks both that the $_SESSION var exists AND that it's not 0.

unspammable-iain at iaindooley dot com (2005-12-23 21:57:44)

As we all know, if an object is serialised, then the class definition must be included _before_ it is unserialised.
My framework has an enormous number of class files, and including them all at the beginning of the script was really taking it's toll on my system (memory and execution time) so I switched to including required classes at the top of each class file that used them using require_once.
This caused problems because I start my session at the very beginning of my script's execution, but all my class files aren't there at the beginning!!
So no in my special 'require' function, I do the following:
if(!class_exists($to_require))
{
session_write_close();
require_once('path/to/classes/'.$to_require.'.php');
session_start();
}
This is a considerably smaller performance hit that including every class that the application uses at the very beginning of the application.

cenaculo at netcabo dot pt (2005-07-23 03:32:28)

This function is essencial when you change $_SESSION[] variables and then, at some poit in the middle of the script, you send an header("Location: http://...") function to the browser, because in this case the session variables may not be saved before the browser change to the new page.
To prevent from lossing session data, allways use session_write_close before this header function. session_write_close will force session data to be saved before the browser change to the new page.
Hope this will help you not to loose 1 day wondering why people could not authenticate or make other changes in session vars in your site.

bkatz at usefulengineering dot com (2005-07-16 20:05:42)

session_write_close() worked as a lifesaver for me when automatically uploading files to a user (forcing a download instead of a link). If files are large, and since session_start() does not allow another page using session_start() to proceed until it's done, i was not able to upload more than one file at a time. By using session_write_close() before beginning the file upload, my users can now download as many big files as they like, at the same time. Example:

<?
session_start();
/* Do session stuff here; security; logging; etc. */
session_write_close();
/* NOW write out the requested file. */
header("Content-type: audio/x-mpeg"); /* or whatever type */
header("Content-Disposition: attachment; filename=" . $filename); 
header("Content-Length: " . $filesize);
header("Content-Transfer-Encoding: binary\n\n");
header("Pragma: no-cache"); 
header("Expires: 0"); 
$file_contents = file_get_contents($filepath);    
print($file_contents);
?>

editorial at literati dot ca (2005-05-15 13:13:58)

Further to the comment by nakanishi at mailstyle dot com, it appears that calling session_write_close() followed by session_start() causes issues if you have more than one browser window/tab open in the session, and have a large session data array. I have an intermitent (and hard to replicate reliably) issue with session_start() never being called or not returning - the script hangs before the session headers are written. I'm puting this down to trying to be too clever rather than to a bug per se.

kumar mcmillan (2004-05-06 09:54:05)

if you are trying to work with a larger code base meant for a specific application... and it implements some custom session save handlers, it appears there is no way to reset those save handlers back to the default php state if they are getting in your way. my workaround:
session_write_close(); // close the session at the top of the page :)

jp at webgraphe dot com (2003-11-21 15:50:15)

It is a good idea to call session_write_close() before proceeding to a redirection using
header("Location: URL");
exit();
because it ensures the session is updated (in a file or into a database, depending on the handler you're using) BEFORE you redirect the visitor somewhere else.
JP.

nakanishi at mailstyle dot com (2003-01-18 23:04:56)

Make sure that you call session_start() again after session_write_close() if you rely on the SID rewriting. Otherwise it will not be rewritten.

(2002-04-19 19:48:51)

This function is very useful for session objects. The class def'n of an obj needs to be included before the session is started. You can ignore this by closing the session with this function, and then use session_start() to restart it. Do this after including the class definition, but before using the session variable.