Peter, if you're "yes it can" is about the close of browser triggering onsession end, I really don't think so. Like Andrew says, the session variables remain in memory until timed out or forcibly removed, or upon a restart. It's just that the "ticket" held by the browser pointing to that session in memory is removed. When using J2EE sessions, CF changed from using a long-expiring CFID/CFTOKEN cookie to instead using a non-persistent (in browser memory-only) cookie called jsessionid. Being memory only, it therefore is removed from memory when the browser closes---that is, assuming no other instance of the browser is open holding the same cookie. All these are things I explain in more detail in an article I did in the CFDJ back in July 2002: http://cfdj.sys-con.com/read/41646.htm (discussed in the final section), and the point about browser instances is elborated upon in a blog entry I did around the same time, at http://cfmxplus.blogspot.com/2002/06/j2ee-session-variables-i-mentioned-in.h tml. What I say there about Netscape is so for FireFox. As to whether what I said about IE (then 5 or 6) is still true in 7, I can't say (not yet running it.) For the sake of consistency, I should also correct two points Andrew made, when he said closing the browser "will however assign a new session id via CFID & CFToken". There are two problems there: first, again, when using J2EE sessions those cookies are no longer used for session management (though they will still be used for client variables, which is indeed why they are so long-lasting). Anyway, even then it's not that a new jsessionID cookie would be "assigned" on browser close but rather a new value would be assigned (new cookie created) on the next request from that browser when it hit the CF app. Not trying to be a tall poppy. Just saw this bug discussion blow up and thought I'd check in on it. These two points seemed worth clarifying. It is an area of continued confusion.
/Charlie http://www.carehart.org/ <http://www.carehart.org/images/ace.gif> _____ From: cfaussie@googlegroups.com [mailto:[EMAIL PROTECTED] On Behalf Of Peter Tilbrook Sent: Monday, May 14, 2007 7:54 AM To: cfaussie@googlegroups.com Subject: [cfaussie] Re: onSessionEnd Error Yes it can. But you must use J2EE sessions. This is an option in the CF Administrator. Is it foolproof? Seems to work better than any "client side" solution I've encountered in my 12 years CF usage. On 14/05/07, Andrew Scott <[EMAIL PROTECTED]> wrote: Peter, there seems to be something wrong with what you are saying there. Closing a browser can not trigger a session end event. It however will assign a new session id via CFID & CFToken.. but the actuall session variables still exist in the persistant scope until they expire by the server during the timeout. This can be achieved, byt doing a simple test. If you know how to test it. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "cfaussie" group. To post to this group, send email to cfaussie@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/cfaussie?hl=en -~----------~----~----~----~------~----~------~--~---
<<inline: ace.gif>>
