Bug 927591

Fixed in version qemu-kvm-0.12.1.2-2.355.el6_4.5

verify this bug by kernel-2.6.32-358.11.1.el6.x86_64 and qemu-kvm-0.12.1.2-2.355.el6_4.5.x86_64

Components:
host:
# rpm -q qemu-kvm
qemu-kvm-0.12.1.2-2.355.el6_4.5.x86_64
# uname -r
2.6.32-358.11.1.el6.x86_64

Guset:
# uname -r
2.6.32-381.el6.x86_64

steps:
1.Boot Guest w/ e1000/RTL8139 virtual network device:
1.1 # /usr/libexec/qemu-kvm ...  -netdev tap,id=hostnet0,script=/etc/qemu-ifup  -device e1000,netdev=hostnet0,id=rtl1,mac=54:52:1b:36:1a:01  ...

1.2 # /usr/libexec/qemu-kvm ...  -netdev tap,id=hostnet0,script=/etc/qemu-ifup  -device rtl8139,netdev=hostnet0,id=rtl1,mac=54:52:1b:36:1a:01  ...

2.After boot up, inside guest, check the link status:
# cat /sys/class/net/eth2/operstate 
up
3.Under hmp, turn down the network link, and check inside guest again.
(qemu) set_link rtl1 off

guest# cat /sys/class/net/eth2/operstate 
down

4.reboot guest.
4.1 under hmp, via system-reset
(qemu) system_reset 
4.2 reboot guest inside guest.

Result:
Test above steps w/ RTL8139/e1000/virtio-net-pci, after guest boot up again, guest is still unavailable, and  check the status of guest:
guest# cat /sys/class/net/eth2/operstate 
down
 
So, according above, this bug is fixed by qemu-kvm-0.12.1.2-2.355.el6_4.5.x86_64 of RHEL6.4.z host kernel-2.6.32-358.11.1.el6.x86_64.

Also test bug 908077 (load/unload+migration scenarios) with this build provided above. And verified pass.

Steps:

1. Boot guest with e1000 NIC.
/usr/libexec/qemu-kvm -cpu Penryn -m 2048 -smp 2,sockets=1,cores=2,threads=1 -M pc -enable-kvm -name rel6u4 -drive file=/mnt/rhel64cp3.qcow2,if=none,format=qcow2,werror=stop,rerror=stop,cache=none,id=drive-scsi0-disk0 -device virtio-scsi-pci,id=scsi0,addr=0x4 -device scsi-hd,scsi-id=0,lun=0,bus=scsi0.0,drive=drive-scsi0-disk0,id=virtio-disk0,bootindex=1 -nodefaults -nodefconfig -vnc :20 -monitor stdio -netdev tap,id=hostnet0,script=/etc/qemu-ifup -device e1000,netdev=hostnet0,id=rtl1,mac=54:52:1b:36:1a:02 -vga std -boot menu=on 

2. Boot guest on destination host with listening mode.

3. Inside guest:
# while true; do rmmod e1000; modprobe e1000; sleep 2; done

4. Migrate guest to des host.

5. After migration, inside guest, kill the script in step 3. And check network:
# ifconfig eth0
eth0      Link encap:Ethernet  HWaddr 54:52:1B:36:1A:02  
          inet addr:10.66.7.116  Bcast:10.66.7.255  Mask:255.255.252.0
          inet6 addr: fe80::5652:1bff:fe36:1a02/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:491 errors:0 dropped:0 overruns:0 frame:0
          TX packets:79 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:51578 (50.3 KiB)  TX bytes:12054 (11.7 KiB)

# ping 10.66.7.126 -c 10
PING 10.66.7.126 (10.66.7.126) 56(84) bytes of data.
64 bytes from 10.66.7.126: icmp_seq=1 ttl=64 time=0.824 ms
64 bytes from 10.66.7.126: icmp_seq=2 ttl=64 time=0.156 ms
64 bytes from 10.66.7.126: icmp_seq=3 ttl=64 time=0.134 ms
64 bytes from 10.66.7.126: icmp_seq=4 ttl=64 time=0.156 ms
64 bytes from 10.66.7.126: icmp_seq=5 ttl=64 time=0.132 ms
64 bytes from 10.66.7.126: icmp_seq=6 ttl=64 time=0.159 ms
64 bytes from 10.66.7.126: icmp_seq=7 ttl=64 time=0.136 ms
64 bytes from 10.66.7.126: icmp_seq=8 ttl=64 time=0.157 ms
64 bytes from 10.66.7.126: icmp_seq=9 ttl=64 time=0.134 ms
64 bytes from 10.66.7.126: icmp_seq=10 ttl=64 time=0.133 ms

# service network restart
Shutting down interface eth0:  Device state: 3 (disconnected)
                                                           [  OK  ]
Shutting down loopback interface:                          [  OK  ]
Bringing up loopback interface:                            [  OK  ]
Bringing up interface eth0:  Active connection state: activating
Active connection path: /org/freedesktop/NetworkManager/ActiveConnection/6
state: activated
Connection activated
                                                           [  OK  ]

6.Test for 5 times, and this guest's link always works well after load/unload > migration.

Test scenario: system-reset/reboot guest repeatedly with random timing

1. Boot guest with e1000 driver
# /usr/libexec/qemu-kvm -cpu Penryn -m 2048 -smp 2,sockets=1,cores=2,threads=1 -M pc -enable-kvm -name rel6u4 -drive file=/home/rhel64cp3.qcow2,if=none,format=qcow2,werror=stop,rerror=stop,cache=none,id=drive-scsi0-disk0 -device virtio-scsi-pci,id=scsi0,addr=0x4 -device scsi-hd,scsi-id=0,lun=0,bus=scsi0.0,drive=drive-scsi0-disk0,id=virtio-disk0,bootindex=1 -nodefaults -nodefconfig -vnc :20 -monitor stdio -netdev tap,id=hostnet0,script=/etc/qemu-ifup -device e1000,netdev=hostnet0,id=rtl1,mac=54:52:1b:36:1a:02 -vga std -boot menu=on -monitor unix:/tmp/qiguo1,server,nowait

2. Repeatedly resets by using system_rest guest with random timing by script
#cat reboot.sh 
i=1
while [ $i -lt 900000 ]
do
echo "system_reset"|nc -U /tmp/qiguo1
i=$(($i+1))
number=$(($RANDOM%60+15))
echo $number
sleep $number
done

Result: After one day's repeatedly reboot, guest's network works well:
1.send/receive pkg
2.ping out/in
3.reset network service

So this scenario passed by qemu-kvm-0.12.1.2-2.355.el6_4.5.x86_64 in RHEL6.4.z host w/ kernel-2.6.32-358.11.1.el6.x86_64.

Migrate guest between rhel6.3 and rhel6.4.z with load/unload nic driver.

source:
kernel:2.6.32-279.el6.x86_64
qemu-kvm-0.12.1.2-2.295.el6.x86_64

destination:
kernel:2.6.32-358.11.1.el6.x86_64
qemu-img-rhev-0.12.1.2-2.355.el6_4.5.x86_64

guest:
kernel:2.6.32-358.11.1.el6.x86_64

steps:
1 boot up guest on source host with:
/usr/libexec/qemu-kvm \
-name rhel6u4 \
-M rhel6.3.0 \
-enable-kvm \
-m 4096 \
-smp 2,sockets=2,cores=1,threads=1 \
-uuid 4ee77792-380d-e8e6-0d77-eb91fdf84a17 \
-nodefconfig \
-monitor stdio \
-rtc base=localtime,clock=host,driftfix=slew \
-boot menu=on \
-device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 \
-drive file=/mnt/rhel6u4.qcow2,if=none,id=drive-virtio-disk0,format=qcow2,cache=none \
-device virtio-blk-pci,scsi=on,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 \
-drive file=/mnt/boot.iso,if=none,media=cdrom,id=drive-ide0-1-0,readonly=on,format=raw \
-device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0,bootindex=2 \
-netdev tap,id=hostnet1 \
-device rtl8139,netdev=hostnet1,id=net1,mac=52:54:00:01:01:ef,bus=pci.0,addr=0x7 \
-chardev pty,id=charserial0 \
-device isa-serial,chardev=charserial0,id=serial0 \
-device usb-tablet,id=input0 \
-vnc :0 \
-vga cirrus \
-device intel-hda,id=sound0,bus=pci.0 \
-device hda-duplex \
-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 \
-device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x9 \
-chardev socket,id=channel1,path=/tmp/helloworld1,server,nowait \
-device virtserialport,chardev=channel1,name=port1,bus=virtio-serial0.0,id=port1 \

2. Boot guest on destination host with listening mode.

3. Inside guest:
# while true; do rmmod e1000; modprobe e1000; sleep 2; done

4. Migrate guest to des host.

5. After migration, inside guest, kill the script in step 3. And check network:
[root@localhost network-scripts]# ifdown eth2
[root@localhost network-scripts]# ifup eth2

Determining IP information for eth2... done.

6. do ping-pong migration with load/unload nic driver 5 times.

7. repeat step1~6 with rtl8139

result:
guest nic work well , can get ip address.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHSA-2013-0896.html