help with session

[stathisgotsis]

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';
header("Locatio n: test2.php");
exit();
?>

and test2.php:

<?php
session_start() ;
echo $_SESSION['username'];
?>

Any hint appreciated.

[Erwin Moller]

st***********@h otmail.com wrote:

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';
header("Locatio n: test2.php");
exit();
?>

and test2.php:

<?php
session_start() ;
echo $_SESSION['username'];
?>

Any hint appreciated.

Hi,

Looks fine to me.
Is it possible PHP cannot use sessions at all because of a misconfigured
php.ini?
Can PHP read/write to the directory where sessions are stored (often /tmp)?
Did you accidentally set the session_safe_ha ndler to user?

Regards,
Erwin Moller

[David Haynes]

st***********@h otmail.com wrote:

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test'; session_write_c lose(); // <-- add this header("Locatio n: test2.php");
exit();
?>

and test2.php:

<?php
session_start() ;
echo $_SESSION['username'];
?>

Any hint appreciated.

[Erwin Moller]

David Haynes wrote:

st***********@h otmail.com wrote:

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';

session_write_c lose(); // <-- add this

Hi David,

Why should that help?

according to php.net:

session_write_c lose

(PHP 4 >= 4.0.4, PHP 5)
session_write_c lose -- Write session data and end session
Description
void session_write_c lose ( void )

End the current session and store session data.

Session data is usually stored after your script terminated without the need
to call session_write_c lose(), but as session data is locked to prevent
concurrent writes only one script may operate on a session at any time.
When using framesets together with sessions you will experience the frames
loading one by one due to this locking. You can reduce the time needed to
load all the frames by ending the session as soon as all changes to session
variables are done.

I interpret that as a possibility to quickly close the session if you know
that you won't need it for the remainder of the script, so concurent
request on the same session are not put in line to wait for the first
script to finish.

But since the OP is terminating the script test.php by calling exit(); it
won't help. I think.
I am missing something about session_write_c lose() maybe?

Regards,
Erwin Moller

header("Locatio n: test2.php");
exit();
?>

and test2.php:

<?php
session_start() ;
echo $_SESSION['username'];
?>

Any hint appreciated.

[David Haynes]

Erwin Moller wrote:

David Haynes wrote:

st***********@h otmail.com wrote:

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';

session_write_c lose(); // <-- add this

Hi David,

Why should that help?

according to php.net:

>
session_write_c lose

(PHP 4 >= 4.0.4, PHP 5)
session_write_c lose -- Write session data and end session
Description
void session_write_c lose ( void )

End the current session and store session data.

Session data is usually stored after your script terminated without the need
to call session_write_c lose(), but as session data is locked to prevent
concurrent writes only one script may operate on a session at any time.
When using framesets together with sessions you will experience the frames
loading one by one due to this locking. You can reduce the time needed to
load all the frames by ending the session as soon as all changes to session
variables are done.

I interpret that as a possibility to quickly close the session if you know
that you won't need it for the remainder of the script, so concurent
request on the same session are not put in line to wait for the first
script to finish.

But since the OP is terminating the script test.php by calling exit(); it
won't help. I think.
I am missing something about session_write_c lose() maybe?

Regards,
Erwin Moller

I'm a 'belt and suspenders' kind of guy when it comes to this. By using
session_write_c lose() I know that the data has been flushed into the
cookie prior to ending the program. It just removes one more variable in
the mix.

Having said that, the problem's symptoms could also be explained by
having a setup that is blocking cookies. The session_write_c lose() is
low hanging fruit so why not try it first?

-david-

[Erwin Moller]

David Haynes wrote:

Erwin Moller wrote:

David Haynes wrote:

st***********@h otmail.com wrote:
Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';
session_write_c lose(); // <-- add this
Hi David,

Why should that help?

according to php.net:

>>
session_write_c lose

(PHP 4 >= 4.0.4, PHP 5)
session_write_c lose -- Write session data and end session
Description
void session_write_c lose ( void )

End the current session and store session data.

Session data is usually stored after your script terminated without the
need to call session_write_c lose(), but as session data is locked to
prevent concurrent writes only one script may operate on a session at any
time. When using framesets together with sessions you will experience the
frames loading one by one due to this locking. You can reduce the time
needed to load all the frames by ending the session as soon as all
changes to session variables are done.

I interpret that as a possibility to quickly close the session if you
know that you won't need it for the remainder of the script, so concurent
request on the same session are not put in line to wait for the first
script to finish.

But since the OP is terminating the script test.php by calling exit(); it
won't help. I think.
I am missing something about session_write_c lose() maybe?

Regards,
Erwin Moller

Hi David,
I'm a 'belt and suspenders' kind of guy when it comes to this. By using
session_write_c lose() I know that the data has been flushed into the
cookie prior to ending the program. It just removes one more variable in
the mix.

Having said that, the problem's symptoms could also be explained by
having a setup that is blocking cookies. The session_write_c lose() is
low hanging fruit so why not try it first?
Thanks for your reply.
I am also a belt and suspenders kind of programmer, but this functioncall
just doesn't make sense to me in this situation.

But as fas as I understand the matter: It will only write the sessiondata
away and close the session. But both actions will happen when the script
ends.

Of course, it is entirely possible I am missing something.

Anyway, let's wait what the OP has to say about it.

Regards,
Erwin Moller

-david-

[ED]

"Erwin Moller"
<si************ *************** *************** @spamyourself.c om> wrote in
message news:44******** *************** @news.xs4all.nl ...

David Haynes wrote:

Erwin Moller wrote:

David Haynes wrote:

st***********@h otmail.com wrote:
> Hello everyone,
> I am having a brain fart or something, why won't this work?
>
> test.php:
>
> <?php
> session_start() ;
> $_SESSION['username']='test';
session_write_c lose(); // <-- add this

Hi David,

Why should that help?

according to php.net:

>>>
session_write_c lose

(PHP 4 >= 4.0.4, PHP 5)
session_write_c lose -- Write session data and end session
Description
void session_write_c lose ( void )

End the current session and store session data.

Session data is usually stored after your script terminated without the
need to call session_write_c lose(), but as session data is locked to
prevent concurrent writes only one script may operate on a session at
any
time. When using framesets together with sessions you will experience
the
frames loading one by one due to this locking. You can reduce the time
needed to load all the frames by ending the session as soon as all
changes to session variables are done.

I interpret that as a possibility to quickly close the session if you
know that you won't need it for the remainder of the script, so
concurent
request on the same session are not put in line to wait for the first
script to finish.

But since the OP is terminating the script test.php by calling exit();
it
won't help. I think.
I am missing something about session_write_c lose() maybe?

Regards,
Erwin Moller

Hi David,

I'm a 'belt and suspenders' kind of guy when it comes to this. By using
session_write_c lose() I know that the data has been flushed into the
cookie prior to ending the program. It just removes one more variable in
the mix.

Having said that, the problem's symptoms could also be explained by
having a setup that is blocking cookies. The session_write_c lose() is
low hanging fruit so why not try it first?

Thanks for your reply.
I am also a belt and suspenders kind of programmer, but this functioncall
just doesn't make sense to me in this situation.

But as fas as I understand the matter: It will only write the sessiondata
away and close the session. But both actions will happen when the script
ends.

Of course, it is entirely possible I am missing something.

Anyway, let's wait what the OP has to say about it.

Regards,
Erwin Moller

-david-

hi guys,

Actually had the same issue as the op and using session_write_c lose did
indeed solve the problem.

The user contributed notes at:
http://uk.php.net/manual/en/function...rite-close.php
may shed some light here. Basically these suggest that you should call
session_write_c lose prior to the Location header to ensure that the session
file is correctly updated prior to the redirect (otherwise presumably, the
session file could still be locked/being written from the previous page).

cheers,
ED

[R. Rajesh Jeba Anbiah]

Erwin Moller wrote:
<snip>

But since the OP is terminating the script test.php by calling exit(); it
won't help. I think.
I am missing something about session_write_c lose() maybe?

<snip>

session_write_c lose() will be implicitly called when you use
exit(); but as header redirection is above exit() and some time the
session values may not be written to the desk when that page redirect
is accessed.

--
<?php echo 'Just another PHP saint'; ?>
Email: rrjanbiah-at-Y!com Blog: http://rajeshanbiah.blogspot.com/

[Erwin Moller]

R. Rajesh Jeba Anbiah wrote:

Erwin Moller wrote:
<snip>

But since the OP is terminating the script test.php by calling exit(); it
won't help. I think.
I am missing something about session_write_c lose() maybe? <snip>

session_write_c lose() will be implicitly called when you use
exit(); but as header redirection is above exit() and some time the
session values may not be written to the desk when that page redirect
is accessed.

Hi Rajesh,

Aaah. Thanks, but I am still confused. :-/

Can i summarize it as follows?
When the exit() is reached, the session-writing and closing can be suspended
for some time (due to heavy load on the server eg.) and the next request
(originating from the header("Locatio n: ...") is already trying to use that
same session.

But still I do not understand what can go wrong, since as I understand it
the second page is waiting for the first to finish before it can access the
session, so it hangs and waits.
So no problem there as far as I can see.

I would love to hear more opinions because I never use
session_write_c lose(), but never had any sessiontroubles either, not with
default sessionhandling (php/file) and custom (user/database).

So I would love to know if it is a serious improvement, or just a myth.

Can you/anybody shed some light on that?

TIA

Regards,
Erwin Moller

--
<?php echo 'Just another PHP saint'; ?>
Email: rrjanbiah-at-Y!com Blog: http://rajeshanbiah.blogspot.com/

[Colin McKinnon]

st***********@h otmail.com wrote:

Hello everyone,
I am having a brain fart or something, why won't this work?

test.php:

<?php
session_start() ;
$_SESSION['username']='test';
header("Locatio n: test2.php");
exit();
?>

Can you set a cookie in 302 response?

I'd look at the traffic (squid / ethereal / iehttpheaders / firefox
tamperData) and see. Or start the sessions before loaing test.php

C.