Hide Forgot
Description of problem: Migration will never end after hot-unplug virtual nic. The following testing results base on local migration (on the same client) Version-Release number of selected component (if applicable): kernel version- 2.6.32-133.el6.x86_64 qemu-kvm version - qemu-kvm-0.12.1.2-2.158.el6.x86_64 host:AMD How reproducible: 90% Steps to Reproduce: 1.boot guest (A) with this command line - /usr/libexec/qemu-kvm -M rhel6.1.0 -enable-kvm -m 1024 -smp 2,sockets=2,cores=1,threads=1 -name test -uuid ac0c4b71-6c1a-c864-b844-774ff5663b94 -rtc base=localtime -boot c -drive file=/home/A.qcow2,if=none,id=drive-virtio-disk0,format=qcow2,cache=none -device virtio-blk-pci,bus=pci.0,addr=0x7,drive=drive-virtio-disk0,id=virtio-disk0 -netdev tap,id=hostnet0,vhost=on -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:a7:95:ab,bus=pci.0,addr=0x3 -netdev tap,id=hostnet1,vhost=on -device virtio-net-pci,netdev=hostnet1,id=net1,mac=52:54:00:a6:53:ea,bus=pci.0,addr=0x4 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -usb -device usb-tablet,id=input0 -spice port=5910,disable-ticketing -vga qxl -device intel-hda,id=sound0,bus=pci.0,addr=0x5 -device hda-duplex,id=sound0-codec0,bus=sound0.0,cad=0 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 -monitor stdio 2.from monitor (qemu) netdev_del hostnet0 (qemu) device_del net0 3.boot destination (B) with the following command line - /usr/libexec/qemu-kvm -M rhel6.1.0 -enable-kvm -m 1024 -smp 2,sockets=2,cores=1,threads=1 -name test -uuid ac0c4b71-6c1a-c864-b844-774ff5663b94 -rtc base=localtime -boot c -drive file=/home/A.qcow2,if=none,id=drive-virtio-disk0,format=qcow2,cache=none -device virtio-blk-pci,bus=pci.0,addr=0x7,drive=drive-virtio-disk0,id=virtio-disk0 -netdev tap,id=hostnet1,vhost=on -device virtio-net-pci,netdev=hostnet1,id=net1,mac=52:54:00:a6:53:ea,bus=pci.0,addr=0x4 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -usb -device usb-tablet,id=input0 -spice port=5911,disable-ticketing -vga qxl -device intel-hda,id=sound0,bus=pci.0,addr=0x5 -device hda-duplex,id=sound0-codec0,bus=sound0.0,cad=0 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 -monitor stdio -incoming tcp:0:5888 4.In A's monitor - (qemu) migrate -d tcp:0:5888 Actual results: Please pay attention to the value of 'remaining ram',the value will reduce to 88k/72K/64k.But the migration never end. (qemu) info migrate Migration status: active transferred ram: 1198948 kbytes remaining ram: 15900 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1253196 kbytes remaining ram: 11616 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1266776 kbytes remaining ram: 64 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1276808 kbytes remaining ram: 88 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1281676 kbytes remaining ram: 88 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1284816 kbytes remaining ram: 88 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1292208 kbytes remaining ram: 64 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1294180 kbytes remaining ram: 72 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1296168 kbytes remaining ram: 72 kbytes total ram: 1180040 kbytes (qemu) info migrate Migration status: active transferred ram: 1300608 kbytes remaining ram: 88 kbytes total ram: 1180040 kbytes Expected results: Migration can finished successfully. Additional info:
Does this bug still occur? If so, does the problem also occur when managing the migration via libvirt? Does it still happen if vhost is not used? Does it also happen with an emulated nic (e1000, rtl8139) or just with virtio-net-pci? Thanks.
Won't hit his problem on qemu-kvm-0.12.1.2-2.192.el6.x86_64.
According to comment 4, this seems to be fixed already. Re-open if this is not the case.