用qemu做迁移

参考 http://www.linux-kvm.org/page/Migration

同一台机器：

启动原guest：

~/git/qemu/i386-softmmu/qemu-system-i386 -enable-kvm -drive file=ubuntu.qcow2 -boot dc -m 512  -monitor stdio -netdev user,id=usernet -device rtl8139,netdev=usernet

启动目的guest：

~/git/qemu/i386-softmmu/qemu-system-i386 -enable-kvm -drive file=ubuntu.qcow2 -boot dc -m 512  -monitor stdio -netdev user,id=usernet -device rtl8139,netdev=usernet -incoming tcp:0:4444

在原机器的monitor中输入：

migrate -d tcp:0:4444

查看迁移情况：

info migrate

算法

参考 http://www.linux-kvm.org/page/Migration

1. Setup

• Start guest on destination, connect, enable dirty page logging and more

2. Transfer Memory

• Guest continues to run
• Bandwidth limitation (controlled by the user)
• First transfer the whole memory
• Iteratively transfer all dirty pages (pages that were written to by the guest).

3. Stop the guest

• And sync VM image(s) (guest's hard drives).

4. Transfer State

• As fast as possible (no bandwidth limitation)
• All VM devices' state and dirty pages yet to be transferred

5. Continue the guest

• On destination upon success
  • Broadcast "I'm over here" Ethernet packet to announce new location of NIC(s).
• On source upon failure (with one exception).

一些log

htab_save_iterate(): first_pass=1
htab_save_iterate(): first_pass=1
htab_save_iterate(): first_pass=1
htab_save_iterate(): first_pass=0
htab_save_iterate(): first_pass=0
htab_save_iterate(): first_pass=0
htab_save_iterate(): first_pass=0
Hash table migration complete
First pass (3 iterations): 34147 chunks, 34997 valid, 489291 invalid
Middle passes (4 iterations): 0 chunks, 0 valid, 0 invalid
Final transfer: 0 chunks, 0 valid, 0 invalid

Qemu的一些接口

http://www.linux-kvm.org/page/MigrationQemu0.8.2

和例子中的接口是两套。