NetBSD Problem Report #42248

From www@NetBSD.org  Fri Oct 30 09:29:02 2009
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by www.NetBSD.org (Postfix) with ESMTP id 0A78B63BA2A
	for <gnats-bugs@gnats.netbsd.org>; Fri, 30 Oct 2009 09:29:02 +0000 (UTC)
Message-Id: <20091030092901.C8BDE63B906@www.NetBSD.org>
Date: Fri, 30 Oct 2009 09:29:01 +0000 (UTC)
From: tnn@NetBSD.org
Reply-To: tnn@NetBSD.org
To: gnats-bugs@NetBSD.org
Subject: in_cksum: out of data
X-Send-Pr-Version: www-1.0

>Number:         42248
>Category:       kern
>Synopsis:       in_cksum: out of data
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Oct 30 09:30:00 +0000 2009
>Closed-Date:    Sat Aug 29 15:36:18 +0000 2020
>Last-Modified:  Sat Aug 29 15:36:18 +0000 2020
>Originator:     Tobias Nygren
>Release:        5.99.21
>Organization:
>Environment:
NetBSD build.nygren.pp.se 5.99.21 NetBSD 5.99.21 (XEN3_DOM0.build) #0: Mon Oct 26 17:58:05 CET 2009  root@build.nygren.pp.se:/usr/obj/sys/arch/amd64/compile/XEN3_DOM0.build amd6
>Description:
I've been getting these messages in the syslog of my amd64/XEN3_DOM0.
The periodicity and accuracy of the messages when the machine is idle
is sometimes downright creepy since cron(8) isn't even running.
When loaded, the messages are more irregular but still appear.

Obviously this is highly reproducible and I'm willing to add any instrumentation that might help to identify the cause.

Oct 29 17:11:46 build /netbsd: in_cksum: out of data
Oct 29 19:11:46 build /netbsd: in_cksum: out of data
Oct 29 21:11:46 build /netbsd: in_cksum: out of data
Oct 29 23:11:46 build /netbsd: in_cksum: out of data
Oct 30 01:11:46 build /netbsd: in_cksum: out of data
Oct 30 03:11:46 build /netbsd: in_cksum: out of data
Oct 30 05:11:46 build /netbsd: in_cksum: out of data
Oct 30 07:11:46 build /netbsd: in_cksum: out of data
Oct 30 09:11:46 build /netbsd: in_cksum: out of data

>How-To-Repeat:
Run XEN3_DOM0 on amd64. Use bridge(4), tap(4), vlan(4), msk(4) and pf(4).
Run ntpd(8), dhcpd(8), openvpn and transmission-daemon.

>Fix:

>Release-Note:

>Audit-Trail:
From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/42248: in_cksum: out of data
Date: Mon, 14 Dec 2009 02:46:07 +0000

 On Fri, Oct 30, 2009 at 09:30:00AM +0000, tnn@NetBSD.org wrote:
  > Obviously this is highly reproducible and I'm willing to add any
  > instrumentation that might help to identify the cause.
  > 
  > Oct 29 17:11:46 build /netbsd: in_cksum: out of data

 This message is caused by cpu_in_cksum (in netinet/cpu_in_cksum.c)
 running off the end of its mbuf chain.

 Changing the printf to a panic (or breaking in the debugger or
 whatever) and seeing where it's coming from might be illuminating.

 Given how regular it is, it's probably triggered by something external.

  > Run ntpd(8), dhcpd(8), openvpn and transmission-daemon.

 ...although it could also be dhcp.

 -- 
 David A. Holland
 dholland@netbsd.org

State-Changed-From-To: open->closed
State-Changed-By: tnn@NetBSD.org
State-Changed-When: Sat, 29 Aug 2020 15:36:18 +0000
State-Changed-Why:
can't reproduce any longer. I think it is related to pf,
but I have long since switched to npf.


>Unformatted:

Home	PR Database Search