Hi Raymond,
Thanks for that suggestion - I'll definitely look into it. In the meantime what
we've done is to solve our problem in the hardware, by using 2 network cards
(i.e. 2 IP addresses) in situations where dual port 443 usage is required. This
keeps the code a helluva lot cleaner, and hardware is cheap cheap these days.
But I will have a look at the Apache client nonetheless.
Thanks,
Darryl
----- Original Message ----
From: Raymond Kroeker <[EMAIL PROTECTED]>
To: Tomcat Users List <users@tomcat.apache.org>
Sent: Tuesday, September 9, 2008 5:41:20 PM
Subject: Re: Piggybacking HTTP with binary protocol
Darryl,
One thing I might suggest is instead of rolling your own http
client; is to use the http client from apache http://hc.apache.org. I
use it myself for various projects including binary data transfer. It
will let you focus on your protocol instead of having to implement
http.
Raymond
On Mon, Sep 8, 2008 at 06:50, Darryl Pentz <[EMAIL PROTECTED]> wrote:
> Last time I mailed the list, I was inquiring about implementing a custom
> connector or something along those lines to support a binary protocol along
> with HTTP. This approach proved flawed for various reasons, if not virtually
> impossible to do with the connector, processor, handler architecture etc..
>
> So the approach I was wanting to try next was to amend our in-house protocol
> slightly and include initial HTTP headers so as to make use of the standard
> Http11Processor etc and then continue (after initial servlet mapping) with
> the current binary protocol. My question is, is this possible using say a
> POST.... to then continue comms on the input and output streams using binary?
>
> I have tried to implement this approach but any read I do from the input
> stream after the request is forwarded to the servlet service method, throws
> an EOFException, which I haven't quite figured out. I wrote a simple socket
> client that writes the POST, a host header, and a blank line, I then write 2
> int's and a String, but I am unable to read the ints and the String from the
> input stream in the service method without encountering the EOFException.
>
> Does anybody know what might cause this? I can't quite isolate any code
> within Tomcat that might be causing this issue.
>
> - Darryl
>
>
>
>
> ---------------------------------------------------------------------
> To start a new topic, e-mail: users@tomcat.apache.org
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
--
---------------------------------------------------------
Raymond Kroeker
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
