Wireshark mailing list archives
Re: realtime dumpcap capability
From: Dario Lombardo <lomato () gmail com>
Date: Sun, 10 Mar 2019 21:05:48 +0100
Hi Marty Did you try nflog/nfqueue interface? If that is not fast enough (I haven't done any comparison), I'd suggest you to have a look at ntop projects (like n2disk). It basically depends if you want to high speed capture or dissection. If you can capture and analyze later, have a look at this extcap module (ntop peoject) that leverages n2disk for running wireshark on the db it creates. https://github.com/ntop/n2disk/tree/master/wireshark/extcap On Sun, Mar 10, 2019 at 7:36 PM marty leisner <maleisner () gmail com> wrote:
Running on linux, I'm using two sharktap's across the lan/wan ports of a router. I'm running dumpcap into pipes, and reading the pipes. I want the packets being emitted to be close time between the ingress/egress packets -- what I'm seeing is a difference of up to hundreds of milliseconds which is too long for my use. (on a busy lan, it would be hundred of packets difference). I've played with PIPE_READ_TIMEOUT and WRITER_THREAD_TIMEOUT and haven't gotten much improvement (some, not much) Are there good tutorials for pulling packets out of the linux kernel (with or without libpcap) -- or is it UTSL? marty ___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev () wireshark org> Archives: https://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-request () wireshark org ?subject=unsubscribe
-- Naima is online.
___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev () wireshark org> Archives: https://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-request () wireshark org?subject=unsubscribe
Current thread:
- realtime dumpcap capability marty leisner (Mar 10)
- Re: realtime dumpcap capability Dario Lombardo (Mar 10)