# 使用tar包手工安装多重启动的ubuntu ## 方案概述 服务器已经安装了CentOS操作系统，由于不能满足开发需求，准备将服务器转换成Ubuntu Server 16.04 LTS。但是，远程服务器依然想保留CentOS作为测试使用，所以部署双操作系统多重启动方案。 远程服务器安装和直接可以物理接触的桌面系统不同，不方便从光盘镜像开始从头安装。所以规划如下安装方案： * 如果原操作系统占据了整个磁盘，则通过PXE启动到无盘，然后通过resize方法缩小现有文件系统分区（具体方法和文件系统相关） * 空出足够安装新Ubuntu操作系统的分区 * 线下通过kvm或virtualbox这样的全虚拟化安装一个精简的Ubuntu操作系统，然后通过tar打包方式完整备份整个Ubuntu操作系统 * 将备份的Ubuntu操作系统tar包上传，并解压缩到对应服务器分区 * 修订CentOS的grub2配置，加入启动Ubuntu的配置 * 重启操作系统，选择进入Ubuntu > 以上方法避免了在服务器上重新安装Ubnntu的步骤，并且可以作为今后快速部署Ubuntu的方案。 ## 实施步骤 ## 准备 * 检查操作系统分区划分 ``` #fdisk -l /dev/sda WARNING: fdisk GPT support is currently new, and therefore in an experimental phase. Use at your own discretion. Disk /dev/sda: 240.1 GB, 240057409536 bytes, 468862128 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes Disk label type: gpt # Start End Size Type Name 1 2048 8191 3M BIOS boot parti 2 8192 2105343 1G EFI System 3 2105344 106962943 50G Microsoft basic 4 106962944 111157247 2G Microsoft basic 5 111157248 468860927 170.6G Microsoft basic ``` ``` #df -h Filesystem Size Used Avail Use% Mounted on /dev/sda3 50G 3.2G 44G 7% / devtmpfs 126G 0 126G 0% /dev tmpfs 126G 156K 126G 1% /dev/shm tmpfs 126G 996K 126G 1% /run tmpfs 126G 0 126G 0% /sys/fs/cgroup /dev/sda2 976M 160M 750M 18% /boot ``` ``` #cat /etc/fstab LABEL=/boot /boot ext4 defaults 1 2 LABEL=/ / ext4 defaults 1 1 LABEL=SWAP swap swap defaults 0 0 ... ``` 可以看到当前运行的CentOS操作系统使用了`/dev/sda3`分区，`/dev/sda4`分区是swap，而`/dev/sda5`分区虽然划分，但是实际上未被挂载使用。 所以准备删除掉`/dev/sda4`（需要`swapoff /dev/sda4`并注释掉`/etc/fstab`中`swap`配置行）和`/dev/sda5`分区，然后再独立创建`/dev/sda4`分区提供给Ubuntu操作系统使用。 ## 在kvm环境安装Ubuntu系统 ``` virt-install \ --network bridge:virbr0 \ --name ubuntu16 \ --ram=4096 \ --vcpus=2 \ --disk path=/var/lib/libvirt/images/ubuntu16.img,size=20 \ --graphics vnc \ --cdrom=/var/lib/libvirt/images/ubuntu-16.04.3-server-amd64.iso ``` * 整个安装在一个`/dev/vda1`分区 * 使用备份方法参考[通过tar备份和恢复整个Linux操作系统](https://github.com/huataihuang/cloud-atlas-draft/tree/6f3204fffc11cf006abd394631e2598d98b415c3/os/linux/ubuntu/install/backup_and_restore_system_by_tar/README.md) > ``为备份服务器的IP地址，已设置ssh密钥登陆，这样可以免密钥登陆 首先确保能够从虚拟机内部登陆到远程服务器`` ``` tar -cvpz --one-file-system / | ssh "( cat > ubuntu16_backup.tar.gz )" ``` > 使用`tar -cvpz --one-file-system`备份整个操作系统会跳过`/proc`，`/sys`等目录 ## 恢复ubuntu分区 * 登陆到目标需要恢复ubuntu的服务器，首先调整空出`/dev/sda4`（如上所述） * 恢复备份 挂载作为ubuntu恢复分区文件系统 ``` mount /dev/sda4 /media ``` 使用以下命令恢复 ``` sudo tar -xvpzf /root/ubuntu16_backup.tar.gz -C /media --numeric-owner ``` > `-C`参数告诉`tar`命令先变更目录到`/media`再开始解压缩 * 修订恢复的ubuntu分区中的`/etc/fstab`，这里实际上是修订`/media/etc/fstab`，使其能够挂载`/dev/sda4`分区 首先使用`blkid`获取分区磁盘信息 ``` blkid ``` 输出类似 ``` ... /dev/sda4: UUID="79140412-5137-45cc-81a5-58a8a7dac4d9" TYPE="ext4" PARTUUID="a7bfb616-ae18-46a0-9bcc-4207418af8cf" ... ``` 注意：在`/etc/fstab`中使用的是`UUID`（对比了原先的虚拟机中磁盘命名），所以修改`/media/etc/fstab`如下： ``` UUID=79140412-5137-45cc-81a5-58a8a7dac4d9 / ext4 errors=remount-ro 0 1 ``` ### 恢复GRUB > Ubuntu 16.4采用了传统的GRUB，可以直接编辑配置文件`/boot/grub/menu.lst`，比CentOS 7使用的GRUB 2要简便许多。通过在Ubuntu 16.4中添加CentOS分区中的`/boot`启动目录下内核，可以在同一个启动菜单中切换不同操作系统。 这里会提示在`/etc/default/grub`中添加参数，为了能够在串口输出控制台，参考 [SerialConsoleHowto](https://help.ubuntu.com/community/SerialConsoleHowto) > 设置串口配置方法和Ubuntu版本有关，获取版本信息的方法是`lsb_release -a`，可以看到Ubuntu 16.4代码是`xenial` * 增加ubuntu分区的`/etc/init/ttyS0.conf`（实际修改`/media/etc/init/ttyS0.conf`） - 针对Karmic和更新版本的方法如下 ```bash # ttyS0 - getty # # This service maintains a getty on ttyS0 from the point the system is # started until it is shut down again. start on stopped rc RUNLEVEL=[12345] stop on runlevel [!12345] respawn exec /sbin/getty -L 115200 ttyS0 vt102 ``` > 针对`Edgy/Feisty/Jaunty`需要设置`/etc/event.d/ttyS0`，而针对`Darpper`或更古老版本则修改`/etc/inittab`。这里忽略旧版本设置方法，详细请参考[SerialConsoleHowto](https://help.ubuntu.com/community/SerialConsoleHowto) * 要使得系统能够启动，需要恢复grub，这里需要编辑`/etc/default/grub`（实际配置`/media/etc/default/grub`） ``` # If you change this file, run 'update-grub' afterwards to update # /boot/grub/grub.cfg. GRUB_DEFAULT=0 GRUB_TIMEOUT=1 GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian` GRUB_CMDLINE_LINUX="console=tty0 console=ttyS0,115200n8" # Uncomment to disable graphical terminal (grub-pc only) GRUB_TERMINAL=serial GRUB_SERIAL_COMMAND="serial --speed=115200 --unit=0 --word=8 --parity=no --stop=1" # The resolution used on graphical terminal # note that you can use only modes which your graphic card supports via VBE # you can see them in real GRUB with the command `vbeinfo' #GRUB_GFXMODE=640x480 # Uncomment if you don't want GRUB to pass "root=UUID=xxx" parameter to Linux #GRUB_DISABLE_LINUX_UUID=true ``` * 通过chroot方式运行ubuntu的grub配置 ``` sudo -s for f in dev dev/pts proc sys; do mount --bind /$f /media/$f ; done chroot /media ``` > 注意，这里 mount 需要包含 /sys ，否则在后面 `upgrade-grub` 会出现大量的 `device node not found` 报错。参考: [How to restore GRUB after restoring Debian from backup?](https://unix.stackexchange.com/questions/397927/how-to-restore-grub-after-restoring-debian-from-backup) * 手工修改`/boot/grub/menu.lst`设置如下 ``` title Ubuntu 16.04.3 LTS, kernel 4.10.0-28-generic root (hd0) kernel /boot/vmlinuz-4.10.0-28-generic root=UUID=b513f893-96ca-494c-8634-0ce8eb7dc135 ro console=tty0 console=ttyS0,115200n8 initrd /boot/initrd.img-4.10.0-28-generic title Ubuntu 16.04.3 LTS, kernel 4.10.0-28-generic (recovery mode) root (hd0) kernel /boot/vmlinuz-4.10.0-28-generic root=UUID=b513f893-96ca-494c-8634-0ce8eb7dc135 ro single console=tty0 console=ttyS0,115200n8 initrd /boot/initrd.img-4.10.0-28-generic title CentOS (3.10.0-327.el7.x86_64) 7.2 (Paladin) root (hd0) kernel /boot/vmlinuz-3.10.0-327.el7.x86_64 root=UUID=96219e3e-c371-47a9-baaf-cc2b1b9f8b33 ro crashkernel=auto console=tty0 console=ttyS1,115200 initrd /boot/initramfs-3.10.0-327.el7.x86_64.img ``` > 最后一项是原CentOS 7搬迁过来的GRUB配置 * 重新安装一次grub，覆盖原先CentOS的GRUB - 这里参考[RecoveringUbuntuAfterInstallingWindows](https://help.ubuntu.com/community/RecoveringUbuntuAfterInstallingWindows) ``` sudo grub-install /dev/sda ``` * 重启系统，通过Ubuntu BRUB菜单验证切换操作系统 ## 异常排查 * 启动提示`error: file /boot/grub/i386-pc/normal.mod' not found.` 检查分区 ``` grub rescue> ls (hd0) (hd0,gpt4) (hd0,gpt3) (hd0,gpt2) (hd0,gpt1) (hd1) grub rescue> set cmdpath=(hd0) prefix=(hd0,gpt4)/boot/grub root=hd0,gpt4 ``` 尝试启动 - 参考 [ AndersonIncorp/fix.sh](https://gist.github.com/AndersonIncorp/3acb1d657cb5eba285f4fb31f323d1c3) ``` grub rescue> set prefix=(hd0,gpt4)/usr/lib/grub grub rescue> insmod normal ``` 但是依然报错 ``` grub rescue> insmod normal error: file `/usr/lib/grub/i386-pc/normal.mod' not found. ``` 参考 [GRUB rescue problem after deleting Ubuntu partition!](https://askubuntu.com/questions/493826/grub-rescue-problem-after-deleting-ubuntu-partition) 尝试检查文件系统分区 ``` grub rescue> ls (hd0,gpt1) (hd0,gpt1): Filesystem is unknown. grub rescue> ls (hd0,gpt2) (hd0,gpt2): Filesystem is ext2. grub rescue> ls (hd0,gpt3) (hd0,gpt3): Filesystem is ext2. grub rescue> ls (hd0,gpt4) (hd0,gpt4): Filesystem is ext2. ``` 尝试启动grub ``` grub rescue> set root=(hd0,gpt4) grub rescue> set prefix=(hd0,gpt4)/usr/lib/grub grub rescue> insmod normal error: file `/boot/grub/i386-pc/normal.mod' not found. ``` 如果错误，则重新设置`root`和`prefix`，再次尝试上述步骤 ``` grub rescue> set root=(hd0,gpt3) grub rescue> set prefix=(hd0,gpt3)/usr/lib/grub grub rescue> insmod normal ``` 这次成功了！！！ 这个`(hd0,gpt3)`分区是原先`CentOS`的系统分区 继续按照 ``` grub rescue > normal ``` 检查分区UUID（可以看到UUID确实是`CentOS`的系统分区） ``` grub> ls (hd0,gpt3) Partition hd0,gpt3: Filesystem type ext* - Label `/' - Last modification time 2018-01-29 06:43:17 Monday, UUID 96219e3e-c371-47a9-baaf-cc2b1b9f8b33 - Partition start at 1052672KiB - Total size 52428800KiB grub> ``` 设置启动内核及root分区UUID ``` linux /boot/vmlinuz-3.10.0-327.el7.x86_64 root=UUID=96219e3e-c371-47a9-baaf-cc2b1b9f8b33 ro console=tty0 console=ttyS1,115200 ``` 但是这里遇到问题，系统始终报告`error: file`/boot/vmlinuz-3.10.0-327.el7.x86\_64' not found.\` 尝试了多次，突然想起来，原先系统安装时候，独立了一个`/boot`分区，位于`/dev/sda2`而不是这里的`UUID=96219e3e-c371-47a9-baaf-cc2b1b9f8b33`对应的`/dev/sda3`。 所以重新尝试Ubuntu分区 ``` set root=(hd0,gpt4) set prefix=(hd0,gpt4)/usr/lib/grub insmod normal normal ``` 检查分区UUID ``` grub> ls (hd0,gpt4) Partition hd0,gpt4: Filesystem type ext* - Last modification time 2018-01-28 22:46:34 Sunday, UUID 79140412-5137-45cc-81a5-58a8a7dac4d9 - Partition start at 53481472KiB - Total size 180949575.5KiB ``` 继续加载 ``` linux /boot/vmlinuz-4.10.0-28-generic root=UUID=79140412-5137-45cc-81a5-58a8a7dac4d9 ro console=tty0 console=ttyS0,115200n8 ``` 也是报错 ``` error: file `/boot/vmlinuz-4.10.0-28-generic' not found. ``` 参考 [GRUB rescue on legacy BIOS systems](https://www.pcsuggest.com/grub-rescue-legacy-bios/) * 检查所有可用的磁盘分区和文件系统 ``` grub> ls -l ... error: file `/usr/lib/grub/i386-pc/cpio.mod' not found. error: file `/usr/lib/grub/i386-pc/cbfs.mod' not found. error: file `/usr/lib/grub/i386-pc/btrfs.mod' not found. error: file `/usr/lib/grub/i386-pc/bfs.mod' not found. error: file `/usr/lib/grub/i386-pc/afs.mod' not found. error: file `/usr/lib/grub/i386-pc/affs.mod' not found. No known filesystem detected - Sector size 512B - Total size 234431064KiB Partition hd0,gpt4: Filesystem type ext* - Last modification time 2018-01-28 22:46:34 Sunday, UUID 79140412-5137-45cc-81a5-58a8a7dac4d9 - Partition start at 53481472KiB - Total size 180949575.5KiB Partition hd0,gpt3: Filesystem type ext* - Label `/' - Last modification time 2018-01-29 06:43:17 Monday, UUID 96219e3e-c371-47a9-baaf-cc2b1b9f8b33 - Partition start at 1052672KiB - Total size 52428800KiB Partition hd0,gpt2: Filesystem type ext* - Label `/boot' - Last modification time 2018-01-29 02:34:03 Monday, UUID 5f67c5f1-be04-406e-964c-f49f8b8e940c - Partition start at 4096KiB - Total size 1048576KiB Partition hd0,gpt1: No known filesystem detected - Partition start at 1024KiB - Total size 3072KiB Device hd1: No known filesystem detected - Sector size 512B - Total size 390711384KiB ``` * 尝试启动 ``` set root=(hd0,4) linux (hd0,4)/boot/vmlinuz-4.10.0-28-generic root=UUID=b513f893-96ca-494c-8634-0ce8eb7dc135 ro console=tty0 console=ttyS0,115200n8 initrd (hd0,4)/boot/initrd.img-4.10.0-28-generic ``` 依然找不到 * 发现可以通过指定`(hd0,2)`包含CentOS `/boot`分区来启动 ``` grub> set root=(hd0,3) grub> linux (hd0,2)/vmlinuz-3.10.0-327.ali2014.alios7.x86_64 root=UUID=96219e3e-c371-47a9-baaf-cc2b1b9f8b33 ro console=tty0 console=ttyS1,115200 grub> initrd (hd0,2)/initramfs-3.10.0-327.el7.x86_64.img ``` 然后可以尝试启动CentOS系统了 ``` grub> boot ``` ## CentOS启动 CentOS启动 ``` ... [ 185.715802] dracut-initqueue[749]: Warning: dracut-initqueue timeout - starting timeout scripts [ 186.221782] dracut-initqueue[749]: Warning: dracut-initqueue timeout - starting timeout scripts [ 186.727769] dracut-initqueue[749]: Warning: dracut-initqueue timeout - starting timeout scripts [ 187.233784] dracut-initqueue[749]: Warning: dracut-initqueue timeoWarning: /dev/disk/by-uuid/96219e3e-c371-47a9-baaf-cc2b1b9f8b33 does not exist Generating "/run/initramfs/rdsosreport.txt" Entering emergency mode. Exit the shell to continue. Type "journalctl" to view system logs. You might want to save "/run/initramfs/rdsosreport.txt" to a USB stick or /boot after mounting them and attach it to a bug report. dracut:/# ``` ## 重新开始 `reboot`之后重新开始 检查当前配置： ``` grub rescue> set cmdpath=(hd0) prefix=(hd0,gpt3)/boot/grub root=hd0,gpt3 ``` ``` grub rescue> ls (hd0) (hd0,gpt4) (hd0,gpt3) (hd0,gpt2) (hd0,gpt1) (hd1) ``` 加载grub模块 ``` grub rescue> set prefix=(hd0,gpt3)/usr/lib/grub grub rescue> insmod normal grub rescue> normal ``` 进入常规模式后，检查设备uuuid ``` grub> ls -l Device hd0: No known filesystem detected - Sector size 512B - Total size 234431064KiB Partition hd0,gpt4: Filesystem type ext* - Last modification time 2018-01-28 22:46:34 Sunday, UUID 79140412-5137-45cc-81a5-58a8a7dac4d9 - Partition start at 53481472KiB - Total size 180949575.5KiB Partition hd0,gpt3: Filesystem type ext* - Label `/' - Last modification time 2018-01-29 06:43:17 Monday, UUID 96219e3e-c371-47a9-baaf-cc2b1b9f8b33 - Partition start at 1052672KiB - Total size 52428800KiB Partition hd0,gpt2: Filesystem type ext* - Label `/boot' - Last modification time 2018-01-29 02:34:03 Monday, UUID 5f67c5f1-be04-406e-964c-f49f8b8e940c - Partition start at 4096KiB - Total size 1048576KiB Partition hd0,gpt1: No known filesystem detected - Partition start at 1024KiB - Total size 3072KiB Device hd1: No known filesystem detected - Sector size 512B - Total size 390711384KiB ``` ``` grub> linux (hd0,2)/vmlinuz-3.10.0-327.ali2014.alios7.x86_64 root=/dev/sda3 ro console=tty0 console=ttyS1,115200 grub> initrd (hd0,2)/initramfs-3.10.0-327.el7.x86_64.img ``` > 不知道为何Ubuntu使用的是 ttyS1，而不是ttyS0。如果设置了`ttyS0`反而无法显示输出信息 ## 参考 * [AndersonIncorp/fix.sh](https://gist.github.com/AndersonIncorp/3acb1d657cb5eba285f4fb31f323d1c3) * [GRUB rescue problem after deleting Ubuntu partition! \[duplicate\]](https://askubuntu.com/questions/493826/grub-rescue-problem-after-deleting-ubuntu-partition) * [Rescue GRUB when grub.cfg is missing or corrupted](https://www.pcsuggest.com/grub-rescue-legacy-bios/)