All of lore.kernel.org
 help / color / mirror / Atom feed
From: Patrick McHardy <kaber@trash.net>
To: Jan Engelhardt <jengelh@medozas.de>
Cc: pablo@netfilter.org,
	Netfilter Developer Mailing List
	<netfilter-devel@vger.kernel.org>
Subject: Re: nf_ct_ipv6 doesn't like ICMPv6 MTU notices
Date: Thu, 20 May 2010 16:09:05 +0200	[thread overview]
Message-ID: <4BF54281.9020600@trash.net> (raw)
In-Reply-To: <alpine.LSU.2.01.1005192138390.16764@obet.zrqbmnf.qr>

Jan Engelhardt wrote:
> Heya,
> 
> 
> I notice that -m conntrack --ctstate INVALID marks what seems to be a 
> legitimate packet. StrongSWAN sends out fragmented and large packets for 
> PMTUD, and the returning ICMP packet too big from the tunnel provider 
> falls victim to ctstate INVALID.
> 
> Observed on the left side running 2.6.33.x:
> [1424176.051256] [v6-xinv] IN=sit1 OUT=
> MAC=00:24:21:a7:24:e1:00:21:59:c5:58:5f:08:00:45:00:05:14:00:00:40:00:fb:29:3c:6d:d8:42:50:1e:bc:28:59:ca
> TUNNEL=216.66.80.30->188.40.89.202 SRC=2001:0470:1f0a:0a59:0000:0000:0000:0001
> DST=2001:0470:1f0b:0a59:0000:0000:0000:0001 LEN=1280 TC=0 HOPLIMIT=64 FLOWLBL=0
> PROTO=ICMPv6 TYPE=2 CODE=0 [SRC=2001:0470:1f0b:0a59:0000:0000:0000:0001
> DST=2001:0470:1f0b:1129:0000:0000:0000:0005 LEN=1480 TC=0 HOPLIMIT=64 FLOWLBL=0
> FRAG:0 INCOMPLETE ID:0000003a PROTO=UDP SPT=4500 DPT=4500 LEN=2904 ] MTU=1430
> 
> And even on the other side running a 2.6.34-rc1:
> [v6-xinv] IN=sit1 OUT=
> MAC=00:14:4f:e1:d1:25:00:40:48:b1:5d:18:08:00:45:00:05:14:00:00:40:00:f7:29:7c:d8:d8:42:50:1e:86:4c:53:3b
> TUNNEL=216.66.80.30->134.76.83.59 SRC=2001:0470:1f0a:1129:0000:0000:0000:0001
> DST=2001:0470:1f0b:1129:0000:0000:0000:0005 LEN=1280 TC=0 HOPLIMIT=64 FLOWLBL=0
> PROTO=ICMPv6 TYPE=2 CODE=0 [SRC=2001:0470:1f0b:1129:0000:0000:0000:0005
> DST=2001:0470:1f0b:0a59:0000:0000:0000:0001 LEN=1480 TC=0 HOPLIMIT=64 FLOWLBL=0
> FRAG:0 INCOMPLETE ID:00000070 PROTO=UDP SPT=4500 DPT=4500 LEN=2424 ] MTU=1430
> 
> 21:44:07.473237 IP6 2001:470:1f0a:1129::1 > 2001:470:1f0b:1129::5: ICMP6,
> packet too big, mtu 1430, length 1240
> 
> An idea what's up in nf_conntrack_ipv6?

Try #define DEBUG in net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c.

      reply	other threads:[~2010-05-20 14:09 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-05-19 19:50 nf_ct_ipv6 doesn't like ICMPv6 MTU notices Jan Engelhardt
2010-05-20 14:09 ` Patrick McHardy [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=4BF54281.9020600@trash.net \
    --to=kaber@trash.net \
    --cc=jengelh@medozas.de \
    --cc=netfilter-devel@vger.kernel.org \
    --cc=pablo@netfilter.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.