日期:2015/11/12 - 2015/11/23 time 11:32
主机:n35, n36, n86
目的:再探oVirt-转换xen虚拟机为kvm虚拟机并使用virt-v2v来导入到oVirt中
操作内容:
一、资源 1、源:xen 主机 n35 xen server: n35.test vm: t1 2、vm的配置 [root@n35 pvcreate]# cat /etc/xen/t1.cfg kernel = '/usr/lib/xen/boot/pv-grub-x86_64.gz' extra = "(hd0)/boot/grub/menu.lst" memory = "2048" maxmem = 10000 name = "t1" vcpus = "2" maxvcpus = 10 disk = [ "phy:/dev/vg0/t1-disk,xvda1,w", "phy:/dev/vg0/t1-swap,xvdb1,w" ] # Networking vif = [ "bridge=xenbr1, ip=10.50.200.32, mac=00:16:3E:62:0B:FD","bridge=xenbr2, ip=10.0.205.32, mac=00:16:3E:19:FF:C1" ] # Behaviour on_poweroff = 'destroy' on_reboot = 'restart' on_crash = 'restart' 其中: 2个网卡:分别桥接到xenbr1,xenbr2 2个磁盘:1个是系统分区,1个是swap分区,且磁盘是直接使用的lvm卷 3、vm的控制 启动: xl create /etc/xen/t1.cfg 控制台: xl console t1 停止: xl shutdown t1 列出: xl list 4、目标:kvm 主机 n36 kvm server: n36.test 方便起见,先配置ssh到n36的免口令登录 [root@a02 ~]# ssh-keygen [root@a02 ~]# ssh-copy-id 10.50.200.35 [root@n36 kvm]# df -h Filesystem Size Used Avail Use% Mounted on /dev/sda3 338G 3.7G 317G 2% / tmpfs 16G 76K 16G 1% /dev/shm /dev/sda1 194M 33M 152M 18% /boot /dev/mapper/vg0-lv01 1.0T 13G 1012G 2% /data [root@n36 kvm]# lvs LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert lv01 vg0 -wi-ao---- 1.00t 挂载了一个lv分区作为数据盘,用于存放转换后的vm数据 5、v2v转换服务器 v2v server: a02.test [root@a02 ~]# yum install virt-v2v 方便起见,先配置ssh到n36的免口令登录 [root@a02 ~]# ssh-keygen [root@a02 ~]# ssh-copy-id 10.50.200.36 二、调整vm的配置 1、禁用swap分区 注释/etc/fstab/中的swap配置 [root@t1 ~]# sed -i '/swap/d' /etc/fstab 2、调整 /dev/xvda1这类字符,替换为为UUID格式 [root@t1 ~]# df -h Filesystem Size Used Avail Use% Mounted on /dev/xvda1 7.9G 1.5G 6.1G 19% / tmpfs 934M 0 934M 0% /dev/shm [root@t1 ~]# blkid /dev/xvda1 /dev/xvda1: UUID="2aab5775-9c09-4d11-a246-1a01aaade9aa" TYPE="ext4" [root@t1 ~]# vim /etc/fstab UUID="2aab5775-9c09-4d11-a246-1a01aaade9aa" / ext4 defaults 1 1 3、关闭vm [root@t1 ~]# init 0 三、在另外一台主机上,尝试转换磁盘为可引导 [root@n36 kvm]# mkdir /data/kvm/{images,conf} -p [root@n36 kvm]# cd /data/kvm/ 1、创建 引导分区(/boot) 1)生成一个镜像 [root@n36 kvm]# qemu-img create /data/kvm/images/t1-boot.img 100M 2)格式化 boot 分区,从 cylinder 0 开始 [root@n36 kvm]# sfdisk /data/kvm/images/t1-boot.img (略) /data/kvm/images/t1-boot1 :0(这里输入 0 ,选择 cylinder 0,然后一路回车) (略) Do you want to write this to disk? [ynq] y (略)) 3)创建 devicemapper [root@n36 kvm]# kpartx -av /data/kvm/images/t1-boot.img add map loop0p1 (253:2): 0 401624 linear /dev/loop0 1 4)创建 fs [root@n36 kvm]# mke2fs /dev/mapper/loop0p1 5)挂载文件系统 [root@n36 kvm]# mkdir -p /mnt/boot && mount /dev/mapper/loop0p1 /mnt/boot/ [root@n36 kvm]# df -h /mnt/boot/ Filesystem Size Used Avail Use% Mounted on /dev/mapper/loop0p1 92M 1.6M 85M 2% /mnt/boot [root@n36 kvm]# ls /mnt/boot/ lost+found 6)同步本机的 /boot 目录下的文件 到 上述 新建的 boot 分区的镜像中,作为新虚拟机的 boot 分区,后续要调整内核参数和数据盘的挂载点配置中的根分区为:UUID=xxxx [root@n36 kvm]# du -sh /boot/ 28M /boot/ 确认磁盘空间够用,开始同步: [root@n36 kvm]# rsync -avP /boot/ /mnt/boot/ 7)调整grub配置 [root@n36 kvm]# echo> /mnt/boot/grub/device.map [root@n36 kvm]# vim /mnt/boot/grub/grub.conf 更改root的UUID为t1对应的UUID kernel /vmlinuz-2.6.32-431.el6.x86_64 ro root=UUID=2aab5775-9c09-4d11-a246-1a01aaade9aa(调整这里,UUID在前面有提到) 8)配置grub,写入启动扇区 [root@n36 kvm]# touch /mnt/boot/kvm-boot-image [root@n36 kvm]# grub --device-map=/dev/null (略) grub> device (hd0) /data/kvm/images/t1-boot.img device (hd0) /data/kvm/images/t1-boot.img grub> find /kvm-boot-image find /kvm-boot-image (hd0,0) grub> root (hd0,0) root (hd0,0) Filesystem type is ext2fs, partition type 0x83 grub> setup (hd0) setup (hd0) Checking if "/boot/grub/stage1" exists... no Checking if "/grub/stage1" exists... yes Checking if "/grub/stage2" exists... yes Checking if "/grub/e2fs_stage1_5" exists... yes Running "embed /grub/e2fs_stage1_5 (hd0)"... failed (this is not fatal) Running "embed /grub/e2fs_stage1_5 (hd0,0)"... failed (this is not fatal) Running "install /grub/stage1 (hd0) /grub/stage2 p /grub/grub.conf "... succeeded Done. grub> quit quit 9)卸载 [root@n36 kvm]# umount /mnt/boot && kpartx -d /data/kvm/images/t1-boot.img loop deleted : /dev/loop0 2、创建 根分区(/) 1)源的数据分布 [root@n35 kvm]# lvs LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert lv01 vg0 -wi-ao---- 1.00t t1-disk vg0 -wi-a----- 8.00g t1-swap vg0 -wi-a----- 2.00g 2)导出数据盘t1-disk [root@n35 ~]# qemu-img convert -O raw /dev/vg0/t1-disk /data/test/t1-disk.img 3)检查上面的这个镜像 [root@n35 ~]# e2fsck /data/kvm/images/t1-disk.img e2fsck 1.41.12 (17-May-2010) Superblock last mount time is in the future. (by less than a day, probably due to the hardware clock being incorrectly set) Fix<y>? yes /data/kvm/images/t1-disk.img: clean, 48297/524288 files, 402393/2097152 blocks 有个时间的问题,修复一下,再次查看: [root@n36 kvm]# e2fsck /data/kvm/images/t1-disk.img e2fsck 1.41.12 (17-May-2010) /data/kvm/images/t1-disk.img: clean, 48297/524288 files, 402393/2097152 blocks 4)同步数据盘到本地 [root@n36 kvm]# rsync -avP 10.50.200.35:/data/test/t1-disk.img /data/kvm/images/ receiving incremental file list t1-disk.img 8589934592 100% 85.42MB/s 0:01:35 (xfer#1, to-check=0/1) sent 30 bytes received 8590983250 bytes 89025733.47 bytes/sec total size is 8589934592 speedup is 1.00 【这个步骤耗时:95s】 5)挂载 [root@n36 kvm]# mkdir -p /mnt/disk && mount -o loop /data/kvm/images/t1-disk.img /mnt/disk 查看: [root@n36 kvm]# df -h /mnt/disk/ Filesystem Size Used Avail Use% Mounted on /data/kvm/images/t1-disk.img 7.9G 1.5G 6.1G 19% /mnt/disk [root@n36 kvm]# ls /mnt/disk/ bin dev home lib64 media opt root selinux sys usr boot etc lib lost+found mnt proc sbin srv tmp var 6)同步内核模块文件 [root@n36 kvm]# rsync -avP /lib/modules/ /mnt/disk/lib/modules/ 7)移除/boot [root@n36 kvm]# echo> /mnt/disk/boot/grub/device.map 8)卸载 [root@n36 kvm]# umount /mnt/disk/ 四、使用 kvm 来启动 vm [root@n36 kvm]# vim conf/t1.xml 1、创建一个kvm配置文件: --------------------------------------- <domain type='kvm'> <name>t1</name> <memory unit='KiB'>4096000</memory> <currentMemory unit='KiB'>2048000</currentMemory> <vcpu placement='static'>2</vcpu> <os> <type arch='x86_64' machine='rhel6.5.0'>hvm</type> <boot dev='hd'/> <boot dev='cdrom'/> </os> <features> <acpi/> <apic/> <pae/> </features> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/libexec/qemu-kvm</emulator> <disk type='file' device='disk'> <driver name='qemu' type='raw' cache='none'/> <source file='/data/kvm/images/t1-boot.img'/> <target dev='vda' bus='virtio'/> </disk> <disk type='file' device='disk'> <driver name='qemu' type='raw' cache='none'/> <source file='/data/kvm/images/t1-disk.img'/> <target dev='vdb' bus='virtio'/> </disk> <controller type='usb' index='0'> </controller> <controller type='ide' index='0'> </controller> <interface type='bridge'> <mac address='00:16:3e:3b:fa:3e'/> <source bridge='ovirtmgmt'/> <model type='virtio'/> </interface> <interface type='bridge'> <mac address='00:16:3e:46:4b:31'/> <source bridge='ovirtwan'/> <model type='virtio'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target type='serial' port='0'/> </console> <input type='tablet' bus='usb'/> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes' listen='0.0.0.0'> <listen type='address' address='0.0.0.0'/> </graphics> <video> <model type='cirrus' vram='9216' heads='1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </video> <memballoon model='virtio'> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </memballoon> </devices> </domain> --------------------------------------- 2、定义并启动 t1 [root@n36 kvm]# virsh define conf/t1.xml Domain t1 defined from conf/t1.xml [root@n36 kvm]# virsh start t1 3、验证是否顺利引导进入系统,通过后,关闭vm [root@n36 kvm]# virsh list --all Id Name State ---------------------------------------------------- - t1 shut off 四、使用v2v转换到ovirt中 1、配置存储池(v2v需要根据 pool 来发现 vm 的磁盘) [root@n36 kvm]# virsh pool-create-as --name p1 --type dir --target /data/kvm/images 2、配置v2v服务器,尝试转换 kvm 为 rhev 格式 1)先查看一下远端的 vm 列表: [root@a02 ~]# virsh -c qemu+ssh://[email protected]/system list --all Id Name State ---------------------------------------------------- - t1 shut off 2)开始转换 [root@a02 ~]# date Fri Nov 20 16:41:31 CST 2015 [root@a02 ~]# virt-v2v -ic qemu+ssh://[email protected]/system \ -o rhev -os 10.50.200.86:/data/ovirt/export --network ovirtmgmt \ t1 t1-boot.img: 100% [================================================================]D 0h00m03s t1-disk.img: 100% [================================================================]D 0h02m17s (半天没任何输出,,等待ing,v2v在后台不知道在干啥呢,或许是在转换磁盘中) 此时,可以切到ovirt导出域所在位置,查看v2v的临时目录是否还存在,如果上述操作完毕,该临时目录内的文件将转移到对应的同级目录 images 中: [root@n86 ~]# cd /data/ovirt/export/06df9082-f3e2-4c80-a18c-7418f7bb008a; \ while true; do sleep 1s; \ if [ -d v2v.YvMfaTjy ]; then echo -n '.'; else echo -e "\nDONE!\n`date`"; break fi done .................................................................................................................................................................. DONE! Fri Nov 20 16:59:24 CST 2015 【这个步骤耗时:17m53s,其中拷贝磁盘耗时:2m20s,转换耗费:15m33s】 3、在ovirt页面导入虚拟机 【这个步骤耗时:150s】 五、QA Q1: v2v执行时报错:virt-v2v: Didn't receive full volume. [root@a02 ~]# virt-v2v -ic qemu+ssh://[email protected]/system -o rhev -os 10.50.200.86:/data/ovirt/export --network ovirtmgmt t1 t1-boot.img: 100% [================================================================]D 0h00m03s virt-v2v: Didn't receive full volume. Received 104857600 of 209715200 bytes. A:从提示可以发现,t1-boot.img 拷贝了 100M,v2v得到的信息是 200M,它是从存储池获取的信息,而此处,存储池得到的是旧的信息,可能是有kvm进程在占用(例如vnc连接到图像界面打开virt-manager界面在管理kvm,此时,不妨退出再打开,然后刷新存储池)。 命令行刷新: [root@n36 kvm]# virsh pool-refresh p1 Pool p1 refreshed ZYXW、参考 1、xl and libvirt https://www.redhat.com/archives/libvir-list/2014-May/msg00928.html 2、Migrate XenServer to RHEV https://access.redhat.com/discussions/451463 3、Ovirt : Convert physical/virtual systems to virtual using virt-p2v && virt-v2v then use it in ovirt DC http://website-humblec.rhcloud.com/convert-physical-virtual-virtual-using-virt-v2v-virt-p2v-kvmovirt/ 4、virt-v2v - Convert a guest to use KVM http://libguestfs.org/virt-v2v.1.html 5、Migrate Xen based VM (domU) to KVM based VM http://wiki.kartbuilding.net/index.php/Migrate_Xen_based_VM_(domU)_to_KVM_based_VM 6、Moving disks from Xen to KVM http://www.zomo.co.uk/2012/04/moving-disks-from-xen-to-kvm/