[qemu] 挂载qcow2文件,qcow2里边还有个lvm

环境:archlinux

背景:在虚拟机里玩dpdk,把挂载HugePage(hugetlbfs)的命令写入fstab的时候,写错了,无法启动,需要把qcow2挂起来改一下。

方法:使用qemu-nbd

来源与参考与鸣谢:http://smilejay.com/2012/11/how-to-mount-a-qcow2-image/

                                http://blog.csdn.net/eyf0917/article/details/38339159 

一:要查看内核参数是否支持了nbd,可是完全不知道上哪里去看内核参数(又不是我自己编译的,囧)(什么开始学习内核知识!!!

  我知道了(10个小时后)。。。 不过好想,并没有神马帮助。。。

[root@dpdk ~]# grep -i NBD /boot/config-3.10.0-327.el7.x86_64 
# CONFIG_BLK_DEV_NBD is not set
[root@dpdk ~]# 

   我有知道了更靠谱的一招(又过去了N天之后。。。)

[tong@T7 dpdk]$ zcat /proc/config.gz  |grep -i nbd
CONFIG_BLK_DEV_NBD=m
[tong@T7 dpdk]$ 

  再修正!(几个小时之后) 以上两个方法,得到的只不过是系统编译时的内核参数状态,传给grub的参数变化并不能体现在这里,也许? 

  以上基于经验未求证,经验出处见(dpdk读文档2)

 

二:关于一,从天而降的答案默认内核参数肯定是个’M‘,所以看看有没有加载,没有就载进去,嗯好棒!

[tong@T7 src]$ lsmod |grep nbd
[tong@T7 src]$ modinfo nbd
filename:       /lib/modules/4.7.1-1-ARCH/kernel/drivers/block/nbd.ko.gz
license:        GPL
description:    Network Block Device
depends:        
intree:         Y
vermagic:       4.7.1-1-ARCH SMP preempt mod_unload modversions 
parm:           nbds_max:number of network block devices to initialize (default: 16) (int)
parm:           max_part:number of partitions per device (default: 0) (int)
[tong@T7 src]$ sudo modprobe nbd max_part=16
[tong@T7 src]$ lsmod |grep nbd
nbd                    20480  0
[tong@T7 src]$ 

三:连接nbd设备

[tong@T7 dpdk]$ sudo qemu-nbd -f qcow2 -c /dev/nbd0 disk.img
[tong@T7 dpdk]$ sudo fdisk -l /dev/nbd0
Disk /dev/nbd0: 10 GiB, 10737418240 bytes, 20971520 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x000a11b2

Device      Boot   Start      End  Sectors  Size Id Type
/dev/nbd0p1 *       2048  1026047  1024000  500M 83 Linux
/dev/nbd0p2      1026048 20971519 19945472  9.5G 8e Linux LVM

然后我们发现,咦,竟然没有创建设备nbd0p1与nbd0p2 。然后机智的我去查看了驱动参数是否设置正确。

[tong@T7 dev]$ cat /sys/module/nbd/parameters/max_part 
31
[tong@T7 dev]$ cat /sys/module/nbd/parameters/nbds_max 
16

虽然我搞不清楚它为什么是31,但是反正不是0我就可以忍了。

然后我又机智的想到了partx,对,这都基于我多年来强大的知识积累,虽然我并不太理解这个命令的机制(早就要你去学习内核拉

也可以用partprobe, 应该说最好用partprobe,如果不行再用partx(仿佛记得partx是比较危险的,有一定可能让内核挂掉。?!)

[tong@T7 dev]$ sudo partx -a /dev/nbd0 
[tong@T7 dev]$ ll /dev/nbd0*
brw-rw---- 1 root disk 43, 0 Sep 24 11:03 /dev/nbd0
brw-rw---- 1 root disk 43, 1 Sep 24 11:45 /dev/nbd0p1
brw-rw---- 1 root disk 43, 2 Sep 24 11:45 /dev/nbd0p2

搞定 : )

四:挂吧! 本来故事应该到此结束,然而不行,我这块盘是lvm,可是我不会挂lvm,所以还要插叙一个挂载lvm

快速学习了一下 http://blog.csdn.net/eyf0917/article/details/38339159

然后,并不行,人家的结果是那样的,我的却是这样的,这让人很尬尴。。。 系统死活就是认不得我的lvm。。。

[tong@T7 dpdk]$ sudo lvdisplay
[tong@T7 dpdk]$ sudo vgdisplay
[tong@T7 dpdk]$ sudo pvdisplay
[tong@T7 dpdk]$ sudo lvscan
[tong@T7 dpdk]$ sudo pvdisplay
[tong@T7 dpdk]$ sudo lvmdiskscan
  /dev/sda1   [      50.00 GiB] 
  /dev/nbd0p1 [     500.00 MiB] 
  /dev/sda2   [     150.00 GiB] 
  /dev/nbd0p2 [       9.51 GiB] LVM physical volume
  /dev/sda3   [     731.51 GiB] 
  /dev/sdb1   [     512.00 MiB] 
  /dev/sdb2   [      14.41 GiB] 
  0 disks
  6 partitions
  0 LVM physical volume whole disks
  1 LVM physical volume
[tong@T7 dpdk]$ sudo pvdisplay
[tong@T7 dpdk]$ sudo pvscan  
 No matching physical volumes found
[tong@T7 dpdk]$ sudo pvdisplay
[tong@T7 dpdk]$ man lvmetad

好吧,我只能自己学习,然后我就读到了 “man lvmetad“,细节你们自己去man把,就是说这个daemon cache了各种metadata在它的cache里面,以上命令全部都在读cache。而我的cache根本没有被更新(也许是因为我的systemd配置?没研究),所以人家好使,我的不好使。手动更新使用命令 ” pvscan  --cache“.

[tong@T7 dpdk]$ systemctl list-unit-files |grep lvm
lvm2-lvmetad.service                       disabled 
lvm2-monitor.service                       disabled 
lvm2-pvscan@.service                       static   
lvm2-lvmetad.socket                        static   
[tong@T7 dpdk]$ 
[tong@T7 dpdk]$ sudo pvscan --cache
[tong@T7 dpdk]$ sudo lvdisplay     
  --- Logical volume ---
  LV Path                /dev/centos/swap
  LV Name                swap
  VG Name                centos
  LV UUID                p49k51-9Z0H-hvwk-mA3u-Hhq7-kKKr-OkeYNG
  LV Write Access        read/write
  LV Creation host, time dpdk, 2016-09-18 22:22:36 +0800
  LV Status              NOT available
  LV Size                1.00 GiB
  Current LE             256
  Segments               1
  Allocation             inherit
  Read ahead sectors     auto
   
  --- Logical volume ---
  LV Path                /dev/centos/root
  LV Name                root
  VG Name                centos
  LV UUID                1pUpjb-du22-XxqQ-M6fE-o0Tz-leQ7-IgsfjS
  LV Write Access        read/write
  LV Creation host, time dpdk, 2016-09-18 22:22:37 +0800
  LV Status              NOT available
  LV Size                8.47 GiB
  Current LE             2168
  Segments               1
  Allocation             inherit
  Read ahead sectors     auto
   
[tong@T7 dpdk]$ sudo vgdisplay
  --- Volume group ---
  VG Name               centos
  System ID             
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  4
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                2
  Open LV               0
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               9.51 GiB
  PE Size               4.00 MiB
  Total PE              2434
  Alloc PE / Size       2424 / 9.47 GiB
  Free  PE / Size       10 / 40.00 MiB
  VG UUID               zt0APb-5ZeI-xyZ5-GTZC-J0GQ-ME4F-c3FqWZ
   
[tong@T7 dpdk]$ sudo pvdisplay
  --- Physical volume ---
  PV Name               /dev/nbd0p2
  VG Name               centos
  PV Size               9.51 GiB / not usable 3.00 MiB
  Allocatable           yes 
  PE Size               4.00 MiB
  Total PE              2434
  Free PE               10
  Allocated PE          2424
  PV UUID               97pf2w-aFZr-MshR-17b2-Is9X-KqOg-mdlQzo
   
[tong@T7 dpdk]$ 
pvscan --cache

嗯,然后我终于跟大家是一样的了,可以继续挂载lvm了。

这个时候,我并没有看见我lvm设备:

[tong@T7 dpdk]$ ll /dev/centos/root
ls: cannot access '/dev/centos/root': No such file or directory
[tong@T7 dpdk]$ ll /dev/centos/swap
ls: cannot access '/dev/centos/swap': No such file or directory
[tong@T7 dpdk]$ 

然后,我也不知道为啥要这么干,反正人家说这么干。。。

[tong@T7 dpdk]$ man vgchange
[tong@T7 dpdk]$ sudo vgchange -a y centos
  2 logical volume(s) in volume group "centos" now active
[tong@T7 dpdk]$ ll /dev/centos/
total 0
lrwxrwxrwx 1 root root 7 Sep 24 13:20 root -> ../dm-1
lrwxrwxrwx 1 root root 7 Sep 24 13:20 swap -> ../dm-0
[tong@T7 dpdk]$ ll /dev/dm-0 
brw-rw---- 1 root disk 254, 0 Sep 24 13:20 /dev/dm-0

五,呃,终于可以挂了!

[tong@T7 dpdk]$ sudo mount /dev/centos/root mnt/
[tong@T7 dpdk]$ cd mnt/
[tong@T7 mnt]$ ls
bin  boot  dev  etc  home  lib  lib64  media  mnt  opt  proc  root  run  sbin  srv  sys  tmp  usr  var
[tong@T7 dpdk]$ sudo vim mnt/etc/fstab 

六,搞定,原路返回。机智的我还参悟了 vgchange -an 这个酷酷的用法!终于,完美退栈!

[tong@T7 dpdk]$ sudo umount mnt/
[tong@T7 dpdk]$ sudo vgchange -an centos
  0 logical volume(s) in volume group "centos" now active
[tong@T7 dpdk]$ sudo ls /dev/centos
ls: cannot access '/dev/centos': No such file or directory
[tong@T7 dpdk]$ sudo qemu-nbd -d /dev/nbd0
/dev/nbd0 disconnected
[tong@T7 dpdk]$ sudo pvdisplay 
  WARNING: Device for PV 97pf2w-aFZr-MshR-17b2-Is9X-KqOg-mdlQzo not found or rejected by a filter.
  WARNING: Device for PV 97pf2w-aFZr-MshR-17b2-Is9X-KqOg-mdlQzo not found or rejected by a filter.
  --- Physical volume ---
  PV Name               [unknown]
  VG Name               centos
  PV Size               9.51 GiB / not usable 3.00 MiB
  Allocatable           yes 
  PE Size               4.00 MiB
  Total PE              2434
  Free PE               10
  Allocated PE          2424
  PV UUID               97pf2w-aFZr-MshR-17b2-Is9X-KqOg-mdlQzo
   
[tong@T7 dpdk]$ sudo pvscan --cache
[tong@T7 dpdk]$ sudo pvdisplay 
[tong@T7 dpdk]$ sudo rmmod nbd           
[tong@T7 dpdk]$ sudo ls /dev/nbd*
ls: cannot access '/dev/nbd*': No such file or directory
[tong@T7 dpdk]$ 

七,全剧终!

 

posted on 2016-09-24 13:38  toong  阅读(2551)  评论(0编辑  收藏  举报