All of lore.kernel.org
 help / color / mirror / Atom feed
From: Frank van Maarseveen <frankvm@frankvm.com>
To: Patrick McHardy <kaber@trash.net>
Cc: linux-kernel@vger.kernel.org,
	Kernel Netdev Mailing List <netdev@vger.kernel.org>
Subject: Re: 2.6.17-rc4: netfilter LOG messages truncated via NETCONSOLE
Date: Wed, 31 May 2006 18:06:11 +0200	[thread overview]
Message-ID: <20060531160611.GA25637@janus> (raw)
In-Reply-To: <447DAEC9.3050003@trash.net>

On Wed, May 31, 2006 at 04:57:13PM +0200, Patrick McHardy wrote:
> Frank van Maarseveen wrote:
> > I have two machines named "porvoo" and "espoo". The first one
> > has netconsole configured to send kernel messages to UDP port 514
> > (a.k.a. syslog) on the other machine.
> > 
> > Somewhere between 2.6.13.2 and 2.6.17-rc4 there is a regression causing
> > the netconsole messages which originate from netfilter to be truncated
> > right after the MAC addresses. For example, /var/log/messages on the
> > sending machine says:
> > 
> > May 31 09:28:11 porvoo kernel: IN=eth0 OUT= MAC=00:12:3f:85:9f:92:00:04:9a:a0:1d:d1:08:00 SRC=192.168.100.30 DST=172.17.1.113 LEN=60 TOS=0x00 PREC=0x00 TTL=54 ID=51496 DF PROTO=TCP SPT=50868 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 
> > 
> > but netconsole messages captured in /var/log/messages on the receiving
> > machine:
> > 
> > May 31 09:28:11 porvoo IN=eth0 OUT= 
> > May 31 09:28:11 porvoo MAC=
> > May 31 09:28:11 porvoo 00:
> > May 31 09:28:11 porvoo 12:
> > May 31 09:28:11 porvoo 3f:
> > May 31 09:28:11 porvoo 85:
> > May 31 09:28:11 porvoo 9f:
> > May 31 09:28:11 porvoo 92:
> > May 31 09:28:11 porvoo 00:
> > May 31 09:28:11 porvoo 04:
> > May 31 09:28:11 porvoo 9a:
> > May 31 09:28:11 porvoo a0:
> > May 31 09:28:11 porvoo 1d:
> > May 31 09:28:11 porvoo d1:
> > May 31 09:28:11 porvoo 08:
> > May 31 09:28:11 porvoo 00 
> > May 31 09:49:06 espoo -- MARK --
> > 
> > I ran a tcpdump on the sending machine to verify(?) what goes out but in
> > that case the 2.6.17-rc4 kernel starts to report "protocol 0000 is buggy":

[...]

> 
> 
> The message means that there was recursion and netpoll fell back
> to dev_queue_xmit This patch should fix the "protocol is buggy"
> messages, netpoll didn't set skb->nh.raw. Please try if it also
> makes the other problem go away.

"protocol 0000 is buggy" is gone. The other problem is still there.

-- 
Frank

  reply	other threads:[~2006-05-31 16:06 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-31  9:46 2.6.17-rc4: netfilter LOG messages truncated via NETCONSOLE Frank van Maarseveen
2006-05-31 14:57 ` Patrick McHardy
2006-05-31 16:06   ` Frank van Maarseveen [this message]
2006-05-31 16:36     ` Patrick McHardy
2006-05-31 17:29       ` Frank van Maarseveen
2006-05-31 17:46         ` Patrick McHardy
2006-05-31 18:20           ` Frank van Maarseveen
2006-06-01  9:11           ` Frank van Maarseveen
2006-06-01 17:34             ` Patrick McHardy
2006-06-02 12:35               ` Frank van Maarseveen
2006-06-02 14:02                 ` 2.6.17-rc4: netfilter LOG messages truncated via NETCONSOLE (2) Frank van Maarseveen
2006-06-02 14:16                   ` Patrick McHardy
2006-06-02 14:39                     ` Frank van Maarseveen

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=20060531160611.GA25637@janus \
    --to=frankvm@frankvm.com \
    --cc=kaber@trash.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@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 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.