Description of problem: Hot-unplug virtio disk from Src guest, then do migration, core dumped happens on src guest during migration, and Dst guest status is paused. Core dumped is not 100% reproduce, but even migration complete from Src guest, status of Dst guest is still paused. Version-Release number of selected component (if applicable): kvm-83-207.el5 How reproducible: core dumped reproduce once when tried 3 times, paused status happens 100%. Steps to Reproduce: 1.Run Src windows xp guest with two virtio disks /usr/libexec/qemu-kvm -M rhel5.6.0 -m 2G -smp 2 -name winxp -uuid 123465d2-2032-848d-bda0-de7adb149048 -boot cdn -drive file=/mnt/winxp.img,if=virtio,boot=on,bus=0,unit=0,format=qcow2,cache=off -net nic,macaddr=54:52:00:27:12:23,vlan=0,model=virtio -net tap,vlan=0,script=/etc/qemu-ifup -serial pty -parallel none -usb -usbdevice tablet -monitor stdio -spice host=0,ic=on,port=5937,disable-ticketing -qxl 1 -drive file=/mnt/data.img,if=virtio,format=qcow2,cache=off 2.Run Dst guest with only one system disk with listening mode on Dst host 3.using info pci/ info block to check disk pci address and status. 4.hot-unplug data disk pci_del pci_addr=6 5. start migration migrate -d tcp:10.66.91.145:5888 Actual results: After step 5, migration start, and src guest become core dumped. Dst guest keep paused. (qemu) info migrate Migration status: active transferred ram: 103960 kbytes remaining ram: 2079296 kbytes total ram: 2183256 kbytes (qemu) QXL: qxl_vm_change_state_handler: running=0 handle_dev_input: stop handle_dev_input: save QXL SAVE -- last_release_offset=2194622380 Segmentation fault (core dumped) [root@dhcp-91-53 mnt]# (gdb) bt #0 0x0000003b6167c2e0 in memcpy () from /lib64/libc.so.6 #1 0x000000000047382d in qemu_put_buffer (f=0x15c2ddd0, buf=0xc89e9000c89f1 <Address 0xc89e9000c89f1 out of bounds>, size=40976) at savevm.c:476 #2 0x000000000041706e in virtio_blk_save (f=0x15c2ddd0, opaque=0x15132010, version_id=3) at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/hw/virtio-blk.c:348 #3 0x00000000004739cc in qemu_savevm_state_complete (f=0x15c2ddd0) at savevm.c:813 #4 0x000000000046bae0 in migrate_fd_put_ready (opaque=<value optimized out>) at migration.c:263 #5 0x00000000004071ec in qemu_run_timers (ptimer_head=0xb3a360, current_time=70973021) at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/vl.c:1302 #6 0x0000000000409657 in main_loop_wait (timeout=<value optimized out>) at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/vl.c:4081 #7 0x000000000050023a in kvm_main_loop () at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/qemu-kvm.c:596 #8 0x000000000040e6b5 in main_loop (argc=34, argv=0x7fff475a0618, envp=<value optimized out>) at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/vl.c:4100 #9 main (argc=34, argv=0x7fff475a0618, envp=<value optimized out>) at /usr/src/debug/kvm-83-maint-snapshot-20090205/qemu/vl.c:6536 Check Dst guest status: (qemu) info status VM status: paused Expected results: migration should complete, and dst guest status is running. Additional info:
some info pops up in Dst guest monitor: Unknown savevm section or instance 'virtio-blk' 2 load of migration failed
In RHEL5.x you can't do hot-plug/unplug and then migrate. It is known not to work. RHEL6.0 fixed several bugs on this area, and it is not going to be fixed on RHEL5.x.