public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Carlos Velasco <lkml@newipnet.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: linux-kernel@vger.kernel.org, linux-net@vger.kernel.org
Subject: Re: Networking messed up, bad checksum, incorrect length
Date: Fri, 27 Oct 2006 05:49:37 +0200	[thread overview]
Message-ID: <454181D1.1040904@newipnet.com> (raw)
In-Reply-To: <E1GdIWp-0002qX-00@gondolin.me.apana.org.au>

Herbert Xu escribió:

> These packets look like normal TSO packets.  Linux will send a
> packet containing more data than fits in a packet to the NIC.  The
> NIC will then segment the packet for us.
> 
> If that is not possible, you can try disabling TSO with ethtool -K.

I see...
I have tracked the change to 2.6.18.
Using kernel 2.6.17.14 and below makes tcpdump/libpcap to display the
real packets.
Could this be a change in the tg3 driver to support TSO committed in 2.6.18?

And a question... how this TSO affects netfilter?

> In order to see what really goes out, you'll need to run a packet
> dump beyond the NIC.

They are here:

Sniffer traces taken through port mirroring in the switch
(ethereal with filter: host flash.cnio.es):
http://www.nimastelecom.com/smtptraces/smtptrace_portmirror.pcap


But still wondering how could this affect Netfilter... my real problem
is that Netfilter drops ACK packets because it doesn't see the
connection established/related.

I will try to do some more research.

Regards,
Carlos Velasco
CCNP & CCDP Cisco Certified Network Professional



  reply	other threads:[~2006-10-27  3:49 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-27  1:53 Networking messed up, bad checksum, incorrect length Carlos Velasco
2006-10-27  2:51 ` Carlos Velasco
2006-10-27  3:37 ` Herbert Xu
2006-10-27  3:49   ` Carlos Velasco [this message]
2006-10-27  4:00     ` Herbert Xu
2006-10-27  4:03       ` Carlos Velasco
2006-10-27  4:31         ` Herbert Xu
2006-10-29 22:40           ` Carlos Velasco

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=454181D1.1040904@newipnet.com \
    --to=lkml@newipnet.com \
    --cc=herbert@gondor.apana.org.au \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-net@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox