Thnaks for your councils, but really to obtain this information (i.e The cookie JSESSIONID in the request is it always valid.) I mostly prefer asking it to Tomcat than to create a probably incomplete mechanism on Apache. (And personnally I think that adding an interface to obtain an information already existing in a java code is much easier than recreating a mechanism in a C project in order to obtain the same information :) )
So, I reformulate my question differently : Do you know the name of the structure/object managing Tomcat's cookies in a cluster? Or how does Tomcat make to know if cookie is valid? Regard, Thomas Quoting Jean-frederic Clere <[EMAIL PROTECTED]>: > [EMAIL PROTECTED] wrote: > > >Quoting Jean-frederic Clere <[EMAIL PROTECTED]>: > > > > > > > >>[EMAIL PROTECTED] wrote: > >> > >> > >> > >>>Hi, > >>> > >>>I suppose the answer to my question will be no, but I need so much that I > >>> > >>> > >>prefer > >> > >> > >>>to ask the question... > >>> > >>>Is it possible for an external program to test if a session exists on the > >>> > >>> > >>server > >> > >> > >>>Tomcat? > >>>In fact, I am working on mod_jk and I would like to send a HTTP request to > >>>Tomcat only if the cookie JSESSIONID linked to this request is a valid > >>> > >>> > >>session > >> > >> > >>>on Tomcat. > >>> > >>> > >>> > >>> > >>I would do this in mod_jk (or mod_proxy): You receive the JSESSIONID > >>from the Tomcat with the AJP response, you just have to store them. > >> > >> > >> > > > >Yes, of course I have already considered this approach but in a context > where > >there are a lot of clients and severals mod_jks, I will need to synchronize > the > >mod_jks (The HTTP requests may be dispatched randomly on the mod_jks) then > >provide memory to store this data and mechanisms to find the stored > JSESSIONID > >whereas all that already exist in Tomcat... > > > > > For the memory handing have a look on the > httpd(httpd-proxy-scoreboard)/modules/mem ;-) > You probably have to create one slot mem per JSESSION and you need > "controled" access. > > Cheers > > Jean-Frederic > > >Thank you for your answer, I wonder whether I will not take a look on the > Tomcat > >mechanism of replication... (Not really delighted by that :[ ) > > > >Regards, > > > >Thomas > > > > > > > >>Cheers > >> > >>Jean-Frederic > >> > >> > >> > >>>This work has been realized in a context where we want to update the > >>> > >>> > >>application > >> > >> > >>>running on Tomcat (and JOnAS for the EJB server), as a result we have to > >>>separate the clients working on the old version from the clients working > on > >>> > >>> > >>the > >> > >> > >>>new one. > >>> > >>>Regards, > >>> > >>>Thomas > >>> > >>> > >>>--------------------------------------------------------------------- > >>>To unsubscribe, e-mail: [EMAIL PROTECTED] > >>>For additional commands, e-mail: [EMAIL PROTECTED] > >>> > >>> > >>> > >>> > >>> > >>> > >>--------------------------------------------------------------------- > >>To unsubscribe, e-mail: [EMAIL PROTECTED] > >>For additional commands, e-mail: [EMAIL PROTECTED] > >> > >> > >> > >> > > > > > > > >--------------------------------------------------------------------- > >To unsubscribe, e-mail: [EMAIL PROTECTED] > >For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
