Posted by amygdala on 08/10/07 18:39

"Michael Fesser" <netizen@gmx.de> schreef in bericht
news:albpb3dnnonfa1bv9g0c82b1mn557ml8ab@4ax.com...
> .oO(amygdala)
>
>>Exactly. To top it off, I implemented this because I assumed that by not
>>doing so, Opera's session might not be commited before the redirect.
>
> Are you sure you really understood how sessions work? The browser's
> only task is to keep the session ID, that's it. The handling of the
> session itself - open, read, write, close - is done on the server.
> Sessions are browser-independent, there's nothing to commit.
>
> Micha

Actually I do understand how sessions work, that's why this whole thing
didn't make any sense to me. Why would it work on other browsers, but not on
Opera.

But when all logic fails, I tend to resort to ad-hoc trial and error,
including these kind of things. But Rik's latest post in which he poses some
possibilities, start to make sense. I'm gonna keep the exit() statement
behind the redirects, but I'm still gonna investigate whether my application
indeed is doing something else after the redirect, that it shouldn't.

[Back to original message]