NetBSD Problem Report #46199
From cheusov@tut.by Thu Mar 15 11:29:55 2012
Return-Path: <cheusov@tut.by>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
by www.NetBSD.org (Postfix) with ESMTP id C6A0263B946
for <gnats-bugs@gnats.netbsd.org>; Thu, 15 Mar 2012 11:29:55 +0000 (UTC)
Message-Id: <s93bony6r0w.fsf@work.imb.invention.com>
Date: Thu, 15 Mar 2012 14:29:51 +0300
From: cheusov@tut.by
To: gnats-bugs@gnats.NetBSD.org
Subject: 6.0_BETA. "qemu -net tap" + bridge(4). Network problems in guests.
X-Send-Pr-Version: 3.95
>Number: 46199
>Category: kern
>Synopsis: 6.0_BETA. "qemu -net tap" + bridge(4). Network problems in guests.
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: kern-bug-people
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Thu Mar 15 11:30:00 +0000 2012
>Closed-Date: Tue Jun 03 17:00:52 +0000 2014
>Last-Modified: Tue Jun 03 17:00:52 +0000 2014
>Originator: Aleksey Cheusov
>Release: NetBSD 6.0_BETA
>Organization:
>Environment:
System: NetBSD work.imb.invention.com 6.0_BETA NetBSD 6.0_BETA (GENERIC) #0: Mon Mar 12 17:25:05 FET 2012 cheusov@work.imb.invention.com:/srv/obj-current/sys/arch/i386/compile/GENERIC i386
Architecture: i386
Machine: i386
>Description:
For testing purposes I use qemu with tap networking.
/etc/rc.conf:
net_interfaces='fxp0 bridge0'
/etc/ifconfig.bridge0:
!ifconfig tap0 create up
!ifconfig tap1 create up
!ifconfig tap2 create up
!ifconfig tap3 create up
!brconfig bridge0 add fxp0 add tap0 add tap1 add tap2 add tap3
Qemu guests are run with the following scripts:
qemu -no-acpi \
-drive file=/srv/qemu/openbsd.img,index=0,media=disk,cache=none \
-m 192 \
-net nic,model=ne2k_pci,macaddr=00:06:80:A3:85:CE \
-net tap,fd=3 -boot c "$@" 3<>/dev/tap1
qemu -no-acpi \
-hda /srv/qemu/dfly.img \
-m 192 \
-net nic,model=ne2k_pci,macaddr=02:07:83:A8:81:FF \
-net tap,fd=3 \
-boot c "$@" 3<>/dev/tap3
I started them one-by-one and they started sucessfully for some time
and are becomes visible in my LAN.
However, after some time of uptime the network in guests(s) gisappeared.
0 cheusov>ssh vm-openbsd uptime
12:54PM up 19 mins, 0 users, load averages: 0.07, 0.10, 0.08
0 cheusov>ssh vm-dfly uptime
ssh: connect to host vm-dfly port 22: Connection timed out
255 cheusov>ssh vm-openbsd uptime
1:02PM up 26 mins, 0 users, load averages: 0.15, 0.11, 0.09
0 cheusov>
I don't know is this a bug in the kernel or in qemu.
Restarting guest OS doesn't help.
But the following solves the problem.
ifconfig tap3 destroy
ifconfig tap3 create up
brconfig bridge0 add tap3
qemu ... 3<>/dev/tap3
>How-To-Repeat:
See above
>Fix:
No idea
>Release-Note:
>Audit-Trail:
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 13:05:38 +0100
would a down/up on tap3 fix the issue ?
Also, displaying the interface queues could help (maybe netstat -id)
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 26 ans d'experience feront toujours la difference
--
From: Aleksey Cheusov <cheusov@tut.by>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@netbsd.org, kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 15:23:10 +0300
> would a down/up on tap3 fix the issue ?
No.
> Also, displaying the interface queues could help (maybe netstat -id)
Name Mtu Network Address Ipkts Ierrs Opkts
Oerrs Colls Drops
tap3 1500 <Link> f2:0b:a4:c7:dd:00 799 0 28722 0 0 138584
tap3 1500 fe80::/64 fe80::f00b:a4ff:f 799 0 28722 0 0 138584
> --
> Manuel Bouyer <bouyer@antioche.eu.org>
> =A0 =A0 NetBSD: 26 ans d'experience feront toujours la difference
> --
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Aleksey Cheusov <cheusov@tut.by>
Cc: gnats-bugs@netbsd.org, kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 13:43:38 +0100
On Thu, Mar 15, 2012 at 03:23:10PM +0300, Aleksey Cheusov wrote:
> > would a down/up on tap3 fix the issue ?
>
> No.
>
> > Also, displaying the interface queues could help (maybe netstat -id)
>
> Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Colls Drops
> tap3 1500 <Link> f2:0b:a4:c7:dd:00 799 0 28722 0 0 138584
> tap3 1500 fe80::/64 fe80::f00b:a4ff:f 799 0 28722 0 0 138584
So, lots of drops. Do you see Ipkts and Opkts moving when it's in this
state ?
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 26 ans d'experience feront toujours la difference
--
From: Aleksey Cheusov <cheusov@tut.by>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 15:52:22 +0300
> So, lots of drops. Do you see Ipkts and Opkts moving when it's in this
> state ?
I'm not sure I understood your question. But Opkts sremains the same,
a number of drops increased significantly, Ipkts increased a little bit.
tap3 1500 <Link> f2:0b:a4:c7:dd:00 1124 0 28722 0 0 201213
tap3 1500 fe80::/64 fe80::f00b:a4ff:f 1124 0 28722 0 0 201213
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org,
cheusov@tut.by
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 15:59:15 +0100
On Thu, Mar 15, 2012 at 12:55:02PM +0000, Aleksey Cheusov wrote:
> I'm not sure I understood your question. But Opkts sremains the same,
> a number of drops increased significantly, Ipkts increased a little bit.
>
> tap3 1500 <Link> f2:0b:a4:c7:dd:00 1124 0 28722 0 0 201213
> tap3 1500 fe80::/64 fe80::f00b:a4ff:f 1124 0 28722 0 0 201213
OK, so this means packets sent to the VM makes it up to the tap interface,
but nothing consumes them. The VM also sends packets, that are properly
handled.
The bug could be in tap(4), or in qemu not reading from the tap(4) descriptor
to get available packets (maybe ktrace could give more details here ?).
The problem doesn't seem to be at the bridge level.
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 26 ans d'experience feront toujours la difference
--
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org,
cheusov@tut.by
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Thu, 15 Mar 2012 16:00:54 +0100
On Thu, Mar 15, 2012 at 12:55:02PM +0000, Aleksey Cheusov wrote:
> I'm not sure I understood your question. But Opkts sremains the same,
> a number of drops increased significantly, Ipkts increased a little bit.
>
> tap3 1500 <Link> f2:0b:a4:c7:dd:00 1124 0 28722 0 0 201213
> tap3 1500 fe80::/64 fe80::f00b:a4ff:f 1124 0 28722 0 0 201213
BTW, could you see if the affected tap device has OACTIVE flag set when
in this state ?
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 26 ans d'experience feront toujours la difference
--
From: Aleksey Cheusov <cheusov@tut.by>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@netbsd.org, kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org
Subject: Re: kern/46199: 6.0_BETA. "qemu -net tap" + bridge(4). Network
problems in guests.
Date: Fri, 23 Mar 2012 18:31:50 +0300
On Thu, Mar 15, 2012 at 6:00 PM, Manuel Bouyer <bouyer@antioche.eu.org> wro=
te:
> On Thu, Mar 15, 2012 at 12:55:02PM +0000, Aleksey Cheusov wrote:
>> =A0I'm not sure I understood your question. But Opkts sremains the same,
>> =A0a number of drops increased significantly, Ipkts increased a little b=
it.
>>
>> =A0tap3 =A01500 =A0<Link> =A0 =A0 =A0 =A0f2:0b:a4:c7:dd:00 =A0 =A0 1124 =
=A0 =A0 0 =A0 =A028722
>> =A0 0 =A0 =A0 0 201213
>> =A0tap3 =A01500 =A0fe80::/64 =A0 =A0 fe80::f00b:a4ff:f =A0 =A0 1124 =A0 =
=A0 0 =A0 =A028722
>> =A0 0 =A0 =A0 0 201213
>
> BTW, could you see if the affected tap device has OACTIVE flag set when
> in this state ?
Yes, OACTIVE flag is set on "dead" tap3.
Also, I've got an idea that my ipf rules may cause this problem.
When I disabled them I could not reproduce the problem,
at least two virtual machines worked for a day without problems.
I'm not sure this is relevant, though.
From: Quentin Garnier <cube@cubidou.net>
To: gnats-bugs@NetBSD.org
Cc:
Subject: re: kern/46199
Date: Sun, 4 May 2014 01:30:48 +0000
Hi,
I believe this is the same issue as kern/47506.
--
Quentin Garnier - cube@cubidou.net
"See the look on my face from staying too long in one place
[...] every time the morning breaks I know I'm closer to falling"
KT Tunstall, Saving My Face, Drastic Fantastic, 2007.
State-Changed-From-To: open->closed
State-Changed-By: cube@NetBSD.org
State-Changed-When: Tue, 03 Jun 2014 17:00:52 +0000
State-Changed-Why:
Fixed and pulled up to -6.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.