Re: [jetty-user] Jetty request recycling
Jesse McConnell <[email protected]> Fri, 21 Sep 2012 10:50:44 -0500
| Newsgroups | gmane.comp.java.jetty.support |
|---|---|
| Message-ID | <CAPHPUsK63sUD7t8KV80F_1KbXsS7t6uRWnFyv_zdbB131u+u_g@mail.gmail.com> |
Keep in mind that the servlet-api does not actually ensure that some of the request information does not change after the startAsync though, you should copy off any information about pathInfo and the like before starting async operations. Some of the information is available in request attributes though, so take a look at the spec if there is particular information you need cheers, jesse -- jesse mcconnell [email protected] On Fri, Sep 21, 2012 at 10:39 AM, Simone Bordet <[email protected]> wrote: > Hi, > > On Fri, Sep 21, 2012 at 3:41 PM, Silvio Bierman > <[email protected]> wrote: >> Hello all, >> >> Can someone sched some light on when and how Jetty recycles request objects? >> Are the objects guaranteed to stay valid until for example a doGet or doPost >> has returned? I seem to have some cases where request information seems to >> change during execution of a doPost. > > Requests are recycled per-connection, after the response is complete. > If you call request.startAsync() and return from doGet() or doPost(), > the request is kept alive until the request is completed (which may > happen a number of invocations to doGet() or doPost() later, depending > what your application does). > > If you have a reproducible test case, file a bugzilla, and we'll look at it. > > Simon > -- > http://cometd.org > http://webtide.com > Developer advice, services and support > from the Jetty & CometD experts. > ---- > Finally, no matter how good the architecture and design are, > to deliver bug-free software with optimal performance and reliability, > the implementation technique must be flawless. Victoria Livschitz > > --------------------------------------------------------------------- > To unsubscribe from this list, please visit: > > http://xircles.codehaus.org/manage_email > > --------------------------------------------------------------------- To unsubscribe from this list, please visit: http://xircles.codehaus.org/manage_email