Re: Session ID

try using $_COOKIE['PHPSESSID'] instead of $PHPSESSID .


HTH



----- Original Message -----
From: "Keith Sellars" <Keith(at)webgraffix.com>
To: <hwg-languages(at)hwg.org>
Sent: Tuesday, September 17, 2002 9:25 PM
Subject: Re: Session ID


> Paul,
>
> You are correct.  I have split the tests into separate scripts.  The first
2
> scripts (cookie1.php and cookie2.php) tests to set the cookies.
>
> indexTEST4.php test to verify that a php session CAN be initiated and then
> echos the session ID to the screen as a way of verifying this.
> indexTEST4.php does NOT set a cookie although I could easily combine the 2
> scripts to check that it will occur at the same time.
>
> Thanks,
> Keith D Sellars
> WebGraffix
> www.webgraffix.com
>
> "Making database sites seem easy"
>
> ----- Original Message -----
> From: "Paul Roberts" <mail(at)paul-roberts.com>
> To: "Keith Sellars" <Keith(at)webgraffix.com>
> Cc: <hwg-languages(at)hwg.org>
> Sent: Tuesday, September 17, 2002 4:00 PM
> Subject: Re: Session ID
>
>
> Hi Keith just tried the url's and
> http://www.schoolpad.net/webadmin/indexTEST4.php
> did not set a cookie it just prints the session id to the screen, and I'm
on
> accept all cookies, medium made no difference.
>
> Paul Roberts
> http://www.paul-roberts.com
> mail(at)paul-roberts.com
> ++++++++++++++++++++++++
> ----- Original Message -----
> From: "Keith Sellars" <Keith(at)webgraffix.com>
> To: <hwg-languages(at)hwg.org>
> Sent: Tuesday, September 17, 2002 5:25 PM
> Subject: Re: Session ID
>
>
> Hank and everyone,
>
> Here is the situation so far regarding this problem:
>
> I have run several scripts to check things (gotten them to run them from
> their location):
>
> 1)
> http://www.schoolpad.net/webadmin/cookie1.php followed by
> http://www.schoolpad.net/webadmin/cookie2.php.  This told me that their
> system was indeed setting cookies.
>
> 2)
> http://www.schoolpad.net/webadmin/indexTEST4.php.  This told me that
> session_start and session_register were indeed working properly.
>
> NOW, here's the dilemma.  When a user logs in to the system, session_start
> and session_register are invoked to assign the user a PHPSESSID, which is
> then stored in the database.  Doing the above (test4.php) DOES indeed show
> that it is possible to assign a Session ID to the user and that cookies
can
> be set (cookie1.php and cookie2.php).  However, when our program is run
from
> their server through a normal login means, for some reason, a PHPSESSID
> appears to NOT be assigned.  Since the script kicks the user out to an
> "error" page if a PHPSESSID is equal to "" (no value), this becomes a
> problem.  This is what I cannot figure out - if I can run the
indexTEST4.php
> script and see that it is indeed setting the PHPSESSID, then why on earth
> would it NOT be working when the program is run from our servers?
>
> Here is the basic sequence of events when a user logs in:
> The user goes to login.php.
> The user types in their password/username.
> The login.php script runs several checks to determine what level of admin
> the user is, and then inserts several values into one of the database
> tables, one value which is $PHPSESSID.
> Upon a successful match of password and username, the user is then
> redirected to webadmin/index.php.
> The webadmin/index.php file calls the following function:
> getLogin($PHPSESSID, $REMOTE_ADDR, $DBName, 1)
> ...from a central PHP file (file that contains most of the common
> functions).
> This central PHP file then runs the getLogin function which runs a
database
> query that contains:
> Sessions WHERE SessionID = '$PHPSESSID'" (thus calling the value that has
> just been written to the database and returning this value to the
> webadmin/index.php page).
> The webadmin/index.php script then checks for the existence of a
$PHPSESSID
> value. Upon finding a value present, it present the appropriate content to
> the user. Upon finding that no $PHPSESSID value exists, the user is
> presented with text that reflects a log in error.
>
> Based on the test scripts I've posted above, and the sequence of events
that
> I describe directly above, what could be the cause of this?  Could it be
> related to the version of browser they are running?  I've asked them to
> upgrade to IE 6.0 (they are running 5.5 right now).
>
> ANY HELP AT ALL would be greatly appreciated.  I've determined WHAT is
> happening to a large degree, just not WHY!!!
>
> Thanks,
> Keith D Sellars
> WebGraffix
> www.webgraffix.com
>
> "Making database sites seem easy"
>
>
>
>
>
>
>

HWG: hwg-languages mailing list archives, maintained by Webmasters @ IWA