Hi again,

On Wed, May 21, 2008 at 7:25 PM, Aaron Turner <[EMAIL PROTECTED]> wrote:
> On Wed, May 21, 2008 at 10:14 AM, Piers O'Hanlon <[EMAIL PROTECTED]> wrote:
>> I guess arbitrary adjustments to fields/bytes (maybe just byte counts
>> within a packet would be nice) -
>
> For example????  Not trying to be difficult, but I'm trying to
> understand the scope of the problem.   Perhaps more importantly, are
> we talking L2-L4 headers or TCP stream payload data?  Do you need the
> payload stream reassembled (TCP segments, IP frags)?
>
> Also, what do you mean by "byte counts within a packet"?  Not sure how
> that is different from the pad/truncate functionality tcprewrite
> already has.
>
Sorry I didn't explain clearly - I meant that for example - I would
like to be able to replace
the 32nd byte in the packet with an arbitrary number - this could
dependent on the value of
another byte (or sequence). Handier still would be if there was a way
(like tshark allows for
display) to access particular bytes within predefined payload type by
name (as opposed to
byte counting to the right byte within a packet). I guess checksum
would need recalculating.
E.g if I've got a bunch of RTP packets stored and I want to, say tweak
some bits in the header or
payload.

>> Maybe it might be possible to utilise
>> wireshark's packet parsers/decoders - then one can build on their library
>> of formats.
>
> That's something I've though about.  Historically, the wireshark
> developers have discouraged that because the API isn't considered
> stable yet.  Maybe with the 1.0 release that's changed.   Having all
> the decoders sounds great, but just how often do you want to go in and
> edit a field tcprewrite doesn't already support?
>
I guess the advantage is having easy access to the application level fields
which tcprewrite doesn't currently support.

>> If some of this is possible then let me know - I may not have fully read all
>> the docs.
>
> Well nothing is impossible... it's just a question of time/effort.
> Making the fragroute code into a library that I could use in
> tcprewrite was 10ish hours of effort.  Very doable for a single
> developer working part time.  The wireshark code base is MUCH larger
> and constantly changing- that makes integration much more challenging
> for me.
>
Sure - I can see that. May be there's a way to plug in their decoders
in without
making your codebase too dependent on their changes - but if as you say their
API changes often then it may be tricky.

> Anyways, right now my big concern with this whole feature is that it's
> really easy to miss the mark (not provide the functionality people
> actually need and provide functionality nobody cares about) and at the
> same time eat a lot of my time which could be better spent on other
> features.  And of course, if there are other tools (like Scapy?) which
> can do the same thing, I'm not sure what the real value is of me
> reinventing the wheel.
>
True. Though manipulating traces for playback isn't so available.

> Like one feature I'm thinking about is a graphical wizard (most likely
> a webapp since I don't know QT/GTK/Fox or any other widget library and
> my UI skills frankly suck ass) which basically walks people through a
> bunch of questions and spits out a config file for
> tcpprep/tcprewrite/tcpreplay.   Based on most of the questions on this
> list, I think such a tool would be really helpful for a lot of people
> and make tcpreplay useful to a wider userbase.
>
Maybe - I guess the kind of people who want to replay tcpdump streams
tend to know a bit more about what they're doing than the average user.

Thanks again,

Piers.

> Anyways, thanks for the feedback Piers!
>
> --
> Aaron Turner
> http://synfin.net/
> http://tcpreplay.synfin.net/ - Pcap editing & replay tools for Unix
> They that can give up essential liberty to obtain a little temporary
> safety deserve neither liberty nor safety. -- Benjamin Franklin
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Tcpreplay-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
> Support Information: http://tcpreplay.synfin.net/trac/wiki/Support
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Tcpreplay-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
Support Information: http://tcpreplay.synfin.net/trac/wiki/Support

Reply via email to