Bug 1038343

Summary:	F20 virt-manager host hangs or panics when staring a guest with network set to virtio: macvtap as bridge
Product:	[Fedora] Fedora	Reporter:	Reartes Guillermo <rtguille>
Component:	qemu	Assignee:	Fedora Virtualization Maintainers <virt-maint>
Status:	CLOSED DUPLICATE	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	high	Docs Contact:
Priority:	unspecified
Version:	20	CC:	amit.shah, berrange, cfergeau, dwmw2, itamar, laine, pbonzini, rjones, scottt.tw, virt-maint
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-12-05 10:50:13 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Reartes Guillermo 2013-12-05 00:06:54 UTC

Description of problem:

I created a new guest, and then set the network to Host Device p5p1 : MACVTAP, 
and Source Mode : Bridge.
Then the host lockups/freezes/panics inmediatelly after starting the guest.

Host serial console output:

dic 04 20:35:15 stark.espada libvirtd[598]: No se encontraron los metadatos:...e
dic 04 20:35:31 stark.espada kernel: device p5p1 entered promiscuous mode
dic 04 20:35:31 stark.espada NetworkManager[572]: <info> (macvtap0): link con...
dic 04 20:35:31 stark.espada NetworkManager[572]: <info> (macvtap0): carrier ...
dic 04 20:35:31 stark.espada NetworkManager[572]: <info> (macvtap0): new Macv...
dic 04 20:35:31 stark.espada NetworkManager[572]: <info> (macvtap0): exported...
dic 04 20:35:31 stark.espada NetworkManager[572]: <info> (macvtap0): No exist...
dic 04 20:35:31 stark.espada kernel: tun: Universal TUN/TAP device driver, 1.6
dic 04 20:35:31 stark.espada kernel: tun: (C) 1999-2004 Max Krasnyansky <ma...m>
dic 04 20:35:31 stark.espada dbus-daemon[503]: dbus[503]: [system] Activatin...'
dic 04 20:35:31 stark.espada dbus[503]: [system] Activating via systemd: ser...'
dic 04 20:35:31 stark.espada systemd[1]: Starting Virtual Machine and Contai....
dic 04 20:35:31 stark.espada systemd[1]: Created slice Virtual Machine and C....
dic 04 20:35:31 stark.espada systemd[1]: Starting Virtual Machine and Contai....
dic 04 20:35:31 stark.espada dbus-daemon[503]: dbus[503]: [system] Successfu...'
dic 04 20:35:31 stark.espada dbus[503]: [system] Successfully activated serv...'
dic 04 20:35:31 stark.espada systemd[1]: Started Virtual Machine and Contain....
dic 04 20:35:31 stark.espada systemd[1]: Starting Virtual Machine qemu-FNEXT....
dic 04 20:35:31 stark.espada systemd-machined[1607]: New machine qemu-FNEXT-T...
dic 04 20:35:31 stark.espada systemd[1]: Started Virtual Machine qemu-FNEXT-....
[  660.131528] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  664.121017] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  664.145031] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  688.141984] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  692.131474] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  692.155486] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  697.495477] INFO: rcu_sched self-detected stall on CPU[  697.497479] INFO: rcu_sched detected stalls on CPUs/tasks:
[  724.143437] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  724.155442] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  724.167449] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  752.153910] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  752.165915] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  752.177924] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  780.164386] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  780.176391] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  780.188399] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  808.174864] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  808.186869] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  808.198876] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  836.185343] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  836.197348] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  836.209356] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  864.195824] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  864.207828] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  864.219836] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[  877.567827] INFO: rcu_sched self-detected stall on CPU[  877.569828] INFO: rcu_sched detected stalls on CPUs/tasks:
[  904.210796] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[  904.222801] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[  904.234809] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[  932.221277] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[  932.233282] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[  932.245290] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[  960.231758] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[  960.243763] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[  960.255771] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[  988.242240] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[  988.254245] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[  988.266253] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[ 1016.252722] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[ 1016.264726] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[ 1016.276734] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[ 1044.263203] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[ 1044.275208] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[ 1044.287215] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[ 1057.641209] INFO: rcu_sched self-detected stall on CPU[ 1057.642210] INFO: rcu_sched detected stalls on CPUs/tasks:
[ 1084.278177] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[ 1084.290182] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[ 1084.302189] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[ 1112.288658] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[ 1112.300663] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[ 1112.312671] BUG: soft lockup - CPU#2 stuck for 22s! [migration/2:17]
[ 1140.299140] BUG: soft lockup - CPU#0 stuck for 22s! [qemu-system-x86:1610]
[ 1140.311145] BUG: soft lockup - CPU#1 stuck for 22s! [migration/1:12]
[ 1140.323153] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[ 1168.309622] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[ 1168.321627] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[ 1168.333634] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[ 1196.320103] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[ 1196.332108] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[ 1196.344116] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[ 1224.330585] BUG: soft lockup - CPU#0 stuck for 23s! [qemu-system-x86:1610]
[ 1224.342590] BUG: soft lockup - CPU#1 stuck for 23s! [migration/1:12]
[ 1224.354598] BUG: soft lockup - CPU#2 stuck for 23s! [migration/2:17]
[ 1237.712592] INFO: rcu_sched self-detected stall on CPU[ 1237.714594] INFO: rcu_sched detected stalls on CPUs/tasks:

I rebooted and changed the newtwork back to Virtual Network 'default': NAT
I did not encountered any issue, the guest booted right into Anaconda (F20fTC4).



Version-Release number of selected component (if applicable):
[root@stark ~]# yum list installed '*qemu*'
Loaded plugins: langpacks, refresh-packagekit
Installed Packages
ipxe-roms-qemu.noarch                20130517-3.gitc4bce43.fc20  @fedora  
libvirt-daemon-driver-qemu.x86_64    1.1.3.1-2.fc20              @fedora  
qemu-common.x86_64                   2:1.6.1-2.fc20              @fedora  
qemu-guest-agent.x86_64              2:1.6.1-2.fc20              @anaconda
qemu-img.x86_64                      2:1.6.1-2.fc20              @fedora  
qemu-kvm.x86_64                      2:1.6.1-2.fc20              @fedora  
qemu-system-x86.x86_64               2:1.6.1-2.fc20              @fedora

# yum list installed '*libvirt*'                                                                                                                         
Loaded plugins: langpacks, refresh-packagekit                                                                                                                           
Installed Packages                                                                                                                                                      
libvirt-client.x86_64                   1.1.3.1-2.fc20  @fedora
libvirt-daemon.x86_64                   1.1.3.1-2.fc20  @fedora
libvirt-daemon-config-network.x86_64    1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-interface.x86_64  1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-network.x86_64    1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-nodedev.x86_64    1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-nwfilter.x86_64   1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-qemu.x86_64       1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-secret.x86_64     1.1.3.1-2.fc20  @fedora
libvirt-daemon-driver-storage.x86_64    1.1.3.1-2.fc20  @fedora
libvirt-daemon-kvm.x86_64               1.1.3.1-2.fc20  @fedora
libvirt-glib.x86_64                     0.1.7-2.fc20    @fedora
libvirt-python.x86_64                   1.1.3.1-2.fc20  @fedora
 
[root@stark ~]# uname -r
3.11.10-300.fc20.x86_64


How reproducible:
always

Steps to Reproduce:
1. Start a guest with the network card setup as macvtap: bridge


Actual results:
host either crashes or freezes

Expected results:


Additional info:
In the past i also reported problems with macvtap and bridge.

Comment 1 Reartes Guillermo 2013-12-05 00:27:31 UTC

# lspci -nn | grep Eth
01:09.0 Ethernet controller [0200]: Sundance Technology Inc / IC Plus Corp IC Plus IP100A Integrated 10/100 Ethernet MAC + PHY [13f0:0200] (rev 31)
06:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 06)

I was using 06:00.0 (p5p1) with the guest.

Comment 2 Laine Stump 2013-12-05 10:50:13 UTC

This is the same as Bug 1025770, and effects all kernels since at least 3.11.6 in Fedora 19. As noted in that BZ, I've also tested with the latest rawhide kernel (3.13.0.something) and it's still broken there. It is definitely a problem with the kernel, as the older kernel with newer qemu behaves properly.

Although the other BZ is currently marked as F19, I'm planning to change that, since it is in F19, F20 and rawhide.

*** This bug has been marked as a duplicate of bug 1025770 ***