public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Austin S Hemmelgarn <ahferroin7@gmail.com>
To: Pietro Paolini <pulsarpietro@aol.com>, linux-kernel@vger.kernel.org
Subject: Re: IPIP6 Linux Kernel Implementation
Date: Thu, 07 Nov 2013 10:48:45 -0500	[thread overview]
Message-ID: <527BB65D.1070204@gmail.com> (raw)
In-Reply-To: <8D0A9E071C6D495-108C-DE6F@webmail-m208.sysops.aol.com>

On 2013-11-07 09:54, Pietro Paolini wrote:
> Hello everyone,
> 
> I am writing you regarding the implementation of the IPIP6 tunnel
> contained in the Linux Kernel which I am trying to configure in order
> to obtain a DS-Lite (IPIP6) compliant tunnel on a router running a
> linux OS kernel version 2.6.33.
> 
> This my network topology :
> 
> lan --> tun1 --> eth0
> 
> where the MTU of all links is 1500.
This is probably the cause of the problem, when tunneling without
fragmentation you need to reduce the MTU of the link encapsulated within
the tunnel so that the size of the IP header plus the tunneled packet is
less than or equal to the MTU of the encapsulating link.  For an IPIP6
tunnel over a link with an MTU of 1500, the absolute maximum MTU of the
tunnel itself should be 1440.  If you use iproute2 or a similar program
to set the tunnel MTU to 1440 things should work.

> 
> My problem comes up when the packet size of the incoming IPv4 datagram
> on my LAN interface is close to the MTU (1496 byte) : then the packet
> is "lost" in the file ip6_tunnel.c module at this point :
> 
> linux-2.6.33.5/net/ipv6/ip6_tunnel.c
> function ip6_tnl_xmit2()
> {
> ....
>        mtu = dst_mtu(dst) - sizeof (*ipv6h);
>        if (skb->len > mtu) {              --> 1496 > 1460 <--
>                *pmtu = mtu;
>                err = -EMSGSIZE;
>                goto tx_err_dst_release;
>        }
> ....
> }
> 
> As I understand, I could be wrong of course,  IPv6 fragmentation is not
> handled in that module and packets bigger then the mtu - sizeof(IPv6)
> are discarded; I am wondering if I can find a workaround in order to
> solve this problem as in some way use the IPv6 stack which I assume
> handles fragmentation - I could be again wrong - or implement it in the
> ip6_tunnel.c module.
> 
> I am not an experienced kernel developer then is important for me
> understand if I am in the right direction and Linux kernel really does
> not provide this feature in a IPIP6 tunnel or if I am wrong; I need to
> have that for the implementation of a DS-Lite tunnel which as defined
> in the RFC6333 Section 5.3 requires IPv6 packet fragmentation.
> 
> Thanks a lot,
> Pietro.
> -- 
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/


      reply	other threads:[~2013-11-07 15:48 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-07 14:54 IPIP6 Linux Kernel Implementation Pietro Paolini
2013-11-07 15:48 ` Austin S Hemmelgarn [this message]

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=527BB65D.1070204@gmail.com \
    --to=ahferroin7@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pulsarpietro@aol.com \
    /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