Wireshark · Wireshark-dev: Re: [Wireshark-dev] Adding pcap-ng pipe support to dumpcap

[Jeff Morriss <jeff.morriss.ws@xxxxxxxxx>]

On Thu, Aug 31, 2017 at 12:54 PM, Guy Harris <guy@xxxxxxxxxxxx> wrote:

On Aug 31, 2017, at 3:37 AM, Ed Beroset <beroset@xxxxxxxxxxxxxx> wrote:

> On 08/30/2017 09:31 PM, Guy Harris wrote:
>> On Aug 30, 2017, at 6:00 PM, Ed Beroset <beroset@xxxxxxxxxxxxxx> wrote:
>>> but I can't help but think that the general approach you describe is the better long term strategy.
>> Probably.  It means that the interface between *shark and extcap programs would be different - but, while having extcap programs behave like dumpcap might complicate the extcap programs (although some of the code to do that could be in a library used by dumpcap and by extcap programs), it might simplify the Wireshark capture code path.
>
> I'm not sure that the interface between dumpcap and Wireshark/tshark would need to change to accommodate a wider variety of inputs via pipes.

It wouldn't.

The interface between *extcap programs* and Wireshark/tshark would need to change if we want to have extcap programs work like dumpcap, so that they talk directly to Wireshark/tshark, and write directly to a capture file, rather than talking to dumpcap by sending packets over a pipe.  That was Stephen's suggestion, and I think it's worth considering.

A counter argument to this would be that there are some advantages to not using a (temporary) file as the buffer packets.  The ones I've had in mind for some time are:

* Bug 2234 (filtering tshark captures with read filters (-R) no longer works) - an Known Problem in our release notes since privsep came in.

* Bug 1650 (dumpcap can remove a ring-buffer file before *shark has read it; the resulting packet loss is reasonable but error presented to the user is quite bad).

* Just the general idea of using (slow) files for a buffer.