Linux下的三种快照实现方法

1.使用LVM2工具对逻辑卷进行快照管理（最先选择的快照方式）

优点：独立快照，对数据的保护性最好，各快照间没有牵制关系

缺点：需要对每个快照分配空间，对系统空间浪费大，老的快照很可能由于超过可容修改内容而出错

*该方法只需要升级LVM2到最新，若需要进行快照还原的话，Linux内核保证2.6.33以上

2.使用device-mapper层进行multisnap快照管理（比较底层的快照方式）

优点：关联快照，节省空间

缺点：各快照间具有关联性，只可以删除最老的快照，管理型差，不能进行快照合并（只可以通过dd等方式还原快照内容）

*该方法需要对linux内核打补丁，并重新编译

3.使用LVM2管理device-mapper层的multisnap快照（对用户友好、对底层告诉节省空间）

*该方法需要对2.6.33以上Linux内核打补丁，重新编译，对LVM2工具打补丁重新编译

==========================================================================================================================

通过介绍第三种快照方式来统一

命令使用：

[root@localhost ~]# lvcreate -s --sharedstore mikulas -L 512M -c 64K /dev/vg02/lv01
Logical volume "lv01-shared" created

命令相当简单，但是你在一台普通的LINUX上是不能成功的

环境：

1.       2.6.33及以上内核的LINUX

2.       对该内核进行multisnap补丁

3.       对LVM2工具进行multisnap补丁

具体操作步骤：

1.下载Linux内核，我实现版本内核为2.6.34  www.kernel.com

2.下载 multisnap补丁（我使用的是r18）http://people.redhat.com/mpatocka/patches/kernel/new-snapshots/

3.下载 lvm2 源码（我使用的是lvm2.02.60）

4.下载 lvm2 补丁（使用的是2.02.60的补丁） http://people.redhat.com/mpatocka/patches/userspace/new-snapshots

一.首先对Linux打补丁

1）将patch文件放在Linux内核的根部录下，以以下方式打补丁

patch -p1 < dm-bufio.patch //将所有的文件dapatch

具体顺序：

# new snapshots
dm-multisnap-common.patch
dm-bufio.patch
dm-multisnap-mikulas-headers.patch
dm-multisnap-mikulas-alloc.patch
dm-multisnap-mikulas-blocks.patch
dm-multisnap-mikulas-btree.patch
dm-multisnap-mikulas-commit.patch
dm-multisnap-mikulas-delete.patch
dm-multisnap-mikulas-freelist.patch
dm-multisnap-mikulas-io.patch
dm-multisnap-mikulas-snaps.patch
dm-multisnap-mikulas-common.patch
dm-multisnap-mikulas-config.patch
dm-multisnap-daniel.patch

3）修改一下

4）注意将LANG=C，然后开始编译内核
make menuconfig;
make all ;
make modules_install ;
make install ;
reboot

内核编译完成

2. 目前可以测试device-mapper下的multisnap已经可用了，测试如下

//清空目前的逻辑卷

[root@localhost ~]# dd if=/dev/zero of=/dev/mapper/vg1-ms_base bs=4096 count=1

[root@localhost ~]# dd if=/dev/zero of=/dev/mapper/vg1-ms_snapshot bs=4096 count=1

//加载必须的模块、开机记得加载这两个模块

[root@localhost ~]# modprobe dm-store-mikulas

[root@localhost ~]# modprobe dm-multisnapshot

//准备快照卷

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnapshot /dev/mapper/vg01-ms_base /dev/mapper/vg01-ms_snapshot 32 0 mikulas 0|dmsetup create ms

//> multisnapshot <origin> <COW device> <chunksize><generic args> <shared COW store type><snapshot ids>

//所以32应该是chunksize | 0 应该是 通用参数

> Table line arguments:

> - <origin> : origin device

> - <COW device> : shared COW store device

> - <chunksize> : chunk size in 512b sectors

> - <# generic args> : number of generic arguments

> - <generic args> : generic arguments

> sync-snapshots --- synchronize snapshots according to the list

> preserve-on-error --- halt the origin on error in the snapshot store

> - <shared COW store type> : shared COW store type

> mikulas --- provided by the 'dm-store-mikulas' module

> daniel --- provided by the 'dm-store-daniel' module

> - <# shared COW store args> : number of arguments for shared COW store type

> - <shared COW store args> : shared COW store arguments

> cache-threshold size --- a background write is started

> cache-limit size --- a limit for metadata cache size

> If 'sync-snapshots' was specified:

> - <# snapshot ids> : number of snapshot ids

> - <snapshot ids> : snapshot ids in desired sync order

> Once this is done, you should no longer access /dev/sda and

> /dev/sdb and only use /dev/mapper/ms.

[root@localhost ~]#

[root@localhost ~]# dmsetup status /dev/mapper/ms

0 1048576 multisnapshot 5 0 - 1048576 128 128 0

//拍摄快照

[root@localhost ~]# dmsetup message /dev/mapper/ms 0 create

> DM messages:

> *) multisnapshot

> - create : creates next new snapshot id, reports created id through 'status'

> (the snapshot is created once the multisnapshot is suspended)

> - create_subsnap <snapshot id> : create subsnapshot of specified snapshot

> - delete <snapshot id> : delete the specified snapshot

[root@localhost ~]# dmsetup status /dev/mapper/ms

0 1048576 multisnapshot 5 0 0 1048576 128 128 0

> Status output:

> *) multisnapshot <# output args> <errno> <new snapid>

> <total_sectors> <sectors_allocated> <metadata_sectors>

> <# snapshot ids> <snapshot ids>

>

> Status line output arguments:

> - <# shared COW store output args> : number of output arguments before

> snapshot id list

> - <errno> : error number associated with the first error that occurred in

> the store (e.g. -EIO), 0 means the store is active with no errors

> - <new snapid> : snapshot id that will be used for next snapshot, '-' if

> no snapshot is in the process of being created

> - <total_sectors> : total size of the shared store in 512b sectors

> - <sectors_allocated> : number of sectors allocated for data and metadata

> - <metadata_sectors> : number of sectors allocated for metadata

> - <# snapshot ids> : number of snapshot ids

> - <snapshot ids> : snapshot ids for snapshots in the store

[root@localhost ms]# ll /mnt/ms0

total 16

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155416

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155425

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160124

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160125

drwx------ 2 root root 12288 Aug 24 15:49 lost+found

[root@localhost ms]# dmsetup message /dev/mapper/ms 1 create

[root@localhost ms]# dmsetup status /dev/mapper/ms

0 1048576 multisnapshot 5 0 1 1048576 416 192 1 0

//停用ms再启用ms后，才能查看快照

[root@localhost ms]# dmsetup suspend /dev/mapper/ms

[root@localhost ms]# dmsetup resume /dev/mapper/ms

//attach the snapshot（将快照附加回来，可以查看）

[root@localhost ms]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 1 |dmsetup create ms1

> *) multisnap-snap <origin> <snapshot id>

>

> Table line arguments:

> - <origin> : origin device

> - <snapshot id> : id of the snapshot within the shared store

> *) Delete the snapshot:

> dmsetup message /dev/mapper/ms 0 delete <snapID>

> *) Unload it:

> dmsetup remove ms

> dmsetup remove ms0

//查看快照效果

[root@localhost ms]# mkdir /mnt/ms1

[root@localhost ms]# mount /dev/mapper/ms1 /mnt/ms1

[root@localhost ms]# ll /mnt/ms1

total 18

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155416

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155425

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160124

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160125

-rw-r--r-- 1 root root 11 Aug 24 16:15 20120824161553

-rw-r--r-- 1 root root 11 Aug 24 16:15 20120824161554

drwx------ 2 root root 12288 Aug 24 15:49 lost+found

[root@localhost ms]# ll /mnt/ms0

total 16

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155416

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155425

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160124

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160125

drwx------ 2 root root 12288 Aug 24 15:49 lost+found

[root@localhost ms]# ll /mnt/ms

total 18

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155416

-rw-r--r-- 1 root root 11 Aug 24 15:54 20120824155425

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160124

-rw-r--r-- 1 root root 11 Aug 24 16:01 20120824160125

-rw-r--r-- 1 root root 11 Aug 24 16:15 20120824161553

-rw-r--r-- 1 root root 11 Aug 24 16:15 20120824161554

drwx------ 2 root root 12288 Aug 24 15:49 lost+found

//例子：

1.可以看到，创建时的0还不是很明确意思，但是0就可以了

2.在status时就可以看到快照数目变为3个了，目前快照号有3 0 1 2

[root@localhost ms]# dmsetup message /dev/mapper/ms 0 create

[root@localhost ms]# dmsetup status /dev/mapper/ms

0 1048576 multisnapshot 5 0 3 1048576 736 192 3 0 1 2

[root@localhost ms]# dmsetup suspend /dev/mapper/ms

[root@localhost ms]# dmsetup resume /dev/mapper/ms

[root@localhost ms]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 3 |dmsetup create ms3

[root@localhost ms]# mkdir /mnt/ms3

[root@localhost ms]# mount /dev/mapper/ms3 /mnt/ms3

[root@localhost ms]# ll /mnt/ms3

总计 20

-rw-r--r-- 1 root root 11 08-24 15:54 20120824155416

-rw-r--r-- 1 root root 11 08-24 15:54 20120824155425

-rw-r--r-- 1 root root 11 08-24 16:01 20120824160124

-rw-r--r-- 1 root root 11 08-24 16:01 20120824160125

-rw-r--r-- 1 root root 11 08-24 16:15 20120824161553

-rw-r--r-- 1 root root 11 08-24 16:15 20120824161554

-rw-r--r-- 1 root root 11 08-26 10:17 20120826101049

-rw-r--r-- 1 root root 11 08-26 10:17 20120826101051

drwx------ 2 root root 12288 08-24 15:49 lost+found

//重启后的测试、快照依然可用

[root@localhost ~]# lsmod | grep dm

dm_store_mikulas 40001 0

dm_multisnapshot 17332 1 dm_store_mikulas

dm_multipath 13659 0

scsi_dh 4512 1 dm_multipath

dm_snapshot 24203 3

dm_zero 1095 0

dm_mirror 11238 0

dm_region_hash 6060 1 dm_mirror

dm_log 7551 2 dm_mirror,dm_region_hash

dm_mod 51689 28 dm_store_mikulas,dm_multisnapshot,dm_multipath,dm_snapshot,dm_zero,dm_mirror,dm_log

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnapshot /dev/mapper/vg01-ms_base /dev/mapper/vg01-ms_snapshot 32 0 mikulas 0|dmsetup create ms

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 0 |dmsetup create ms0

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 1 |dmsetup create ms1

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 2 |dmsetup create ms2

[root@localhost ~]# echo 0 `blockdev --getsize /dev/mapper/vg01-ms_base` multisnap-snap /dev/mapper/vg01-ms_base 3 |dmsetup create ms3

[root@localhost ~]# dmsetup ls --target multisnapshot

ms (253, 8)

[root@localhost ~]# dmsetup ls --target multisnap-snap

ms3 (253, 12)

ms2 (253, 11)

ms1 (253, 10)

ms0 (253, 9)

[root@localhost ~]# dmsetup table

ms3: 0 1048576 multisnap-snap 253:6 3

ms2: 0 1048576 multisnap-snap 253:6 2

vg01-lv01-real: 0 1048576 linear 8:17 384

ms1: 0 1048576 multisnap-snap 253:6 1

ms0: 0 1048576 multisnap-snap 253:6 0

ms: 0 1048576 multisnapshot 253:6 253:7 32 0 mikulas 0

vg01-lv01_snap1-cow: 0 24576 linear 8:17 1073536

vg01-ms_snapshot: 0 1048576 linear 8:17 2146688

VolGroup00-LogVol01: 0 4128768 linear 8:2 37552512

vg01-ms_base: 0 1048576 linear 8:17 1098112

VolGroup00-LogVol00: 0 37552128 linear 8:2 384

vg01-lv01: 0 1048576 snapshot-origin 253:2

vg01-lv01_snap1: 0 1048576 snapshot 253:2 253:4 P 8

//快照删除

[root@localhost ~]# dmsetup message /dev/mapper/ms 0 delete 0

[root@localhost ~]# umount /dev/mapper/ms0

[root@localhost ~]# dmsetup remove ms0

[root@localhost ~]# dmsetup message /dev/mapper/ms 0 delete 3

[root@localhost ~]# umount /dev/mapper/ms3

[root@localhost ~]# dmsetup remove ms3

[root@localhost ~]# dmsetup status /dev/mapper/ms

0 1048576 multisnapshot 5 0 - 1048576 512 192 2 1 2

[root@localhost ~]# dmsetup ls --target multisnap-snap

ms2 (253, 11)

ms1 (253, 10)

经测试发现可以删除任何版本的快照

但是重启后，无法将快照原卷和快照卷组合为ms，但是换个名字可以。。。了！！

猜测在重启前将ms关闭，大概就不会出类似问题了

//如何进行快照还原

3.重新编译LVM2先对LVM2打补丁，方式还是和给LINUX打补丁类似

patch -p1 < dm-bufio.patch

补丁顺序

# Tested to apply (and work) on: lvm2.git v2_02_64

lvm-merge-log-print.patch

# Shared snapshots
lvm-refactor-exception-type.patch
lvm-shared-lvcreate.patch
lvm-shared-add-argument-to-vg_add_snapshot.patch
lvm-shared-new-segment-type.patch
lvm-shared-fixed-name.patch
lvm-shared-snapshot-flag.patch
device-mapper-shared-target.patch
lvm-shared-make-pointer-from-origin-to-shared-snapshot.patch
lvm-shared-activate.patch
lvm-shared-handle-activation-error.patch

lvm-shared-snapshot-segment-type.patch
lvm-shared-allow-no-size.patch
lvm-shared-test-free-space-after-setting-segtype.patch
lvm-shared-dev-manager.patch
lvm-shared-create-snapshot.patch
lvm-shared-no-snapshot-of-snapshot.patch
lvm-shared-delete-message.patch
lvm-shared-dont-deactivate-origin-when-deactivating-snapshot.patch
lvm-shared-deactivate-snapshots-with-origin.patch
lvm-shared-remove-snapshots-with-origin.patch
lvm-shared-activate-shared-origin-with-the-snapshot.patch
lvm-shared-dont-shrink-shared-volume.patch
lvm-shared-noexplicit-segment.patch
lvm-dont-merge-shared-store.patch
lvm-last-seg.patch
lvm-shared-resize-virtual-snapshot.patch
lvm-shared-cleanup-on-activate.patch
lvm-shared-status.patch
lvm-shared-status-percent.patch
lvm-shared-snapshot-of-snapshot.patch

# Patches by [email protected]:
lvm-shared-additional-wrappers.patch
lvm-shared-eliminate-shared_snapshot-in-lv.patch
lvm-shared-vorigin-segv-fix.patch
lvm-shared-add-ACCESS_HIDDEN_LV-flag.patch

# by Mikulas
lvm-refactor-chunksize-validation.patch
lvm-shared-big-chunksize.patch
lvm-merge-function-onactivate.patch

lvm-shared-merge.patch
lvm-shared-merge-dont-delete.patch

编译LVM2

aclocal

autoconf

./configure

make

make install

[root@localhost bin]# ll /sbin/lvcreate //此处为刚安装的lvcreate

lrwxrwxrwx 1 root root 3 08-27 00:29 /sbin/lvcreate -> lvm

由于还有一个在/usr/sbin/下，系统本来装的

将/sbin下的占到/usr/sbin下，将原lvm作为lvm_old

4.尝试LVM2管理multisnap

//创建快照卷

[root@localhost ~]# lvcreate -s --sharedstore mikulas -L 512M -c 64K /dev/vg02/lv01

Logical volume "lv01-shared" created

//创建快照

[root@localhost ~]# lvcreate -s -n snap_$(date -d "today" +"%Y%m%s%H%I%S") /dev/vg02/lv01

Logical volume "snap_2012081346132205130145" created

//查看目前卷情况，可以看到加了快照以后，快照卷被占用了0.07%

[root@localhost ~]# lvs

LV VG Attr LSize Origin Snap% Move Log Copy% Convert

LogVol00 VolGroup00 -wi-ao 17.91G

LogVol01 VolGroup00 -wi-ao 1.97G

lv01 vg02 owi-a- 512.00M

lv01-shared vg02 swi--- 512.00M lv01 0.07

snap_2012081346132205130145 vg02 swi-a- 512.00M lv01

[root@localhost ~]# mkfs.ext3 /dev/vg02/lv01

[root@localhost ~]# mkdir /mnt/lv01_2012081346132205130145

//第一个快照是没有文件系统的，所以无法挂载

[root@localhost ~]# mount /dev/vg02/snap_2012081346132205130145 /mnt/lv01_2012081346132205130145

mount: you must specify the filesystem type

[root@localhost ~]# lvcreate -s -n snap_$(date -d "today" +"%Y%m%s%H%I%S") /dev/vg02/lv01

Logical volume "snap_2012081346132657130117" created

[root@localhost ~]# mkdir /mnt/snap_2012081346132657130117

//第二个快照是在创建文件系统后拍摄的，所以可以挂载，但内容为空

[root@localhost lv01]# ll /mnt/snap_2012081346132657130117/

总计 12

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost ~]# mount /dev/vg02/snap_2012081346132657130117 /mnt/snap_2012081346132657130117

[root@localhost ~]# cd /mnt/lv01

[root@localhost lv01]# touch $(date -d "today" +"%Y%m%s%H%I%S")

[root@localhost lv01]# touch $(date -d "today" +"%Y%m%s%H%I%S")

//第三个快照是创建了一些数据后拍摄的，所以可以挂载且有内容

[root@localhost lv01]#lvcreate -s -n snap_$(date -d "today" +"%Y%m%s%H%I%S") /dev/vg02/lv01

Logical volume "snap_2012081346132744130144" created

[root@localhost lv01]# mkdir /mnt/snap_2012081346132744130144

[root@localhost lv01]# mount /dev/vg02/snap_2012081346132744130144 /mnt/snap_2012081346132744130144

[root@localhost lv01]# ll /mnt/snap_2012081346132744130144

总计 12

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132730130130

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132731130131

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost lv01]# touch $(date -d "today" +"%Y%m%s%H%I%S")

[root@localhost lv01]# touch $(date -d "today" +"%Y%m%s%H%I%S")

//目前原卷情况

[root@localhost lv01]# ll

总计 12

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132730130130

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132731130131

-rw-r--r-- 1 root root 0 08-28 13:46 2012081346132801130141

-rw-r--r-- 1 root root 0 08-28 13:46 2012081346132802130142

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost lv01]# lvs

LV VG Attr LSize Origin Snap% Move Log Copy% Convert

LogVol00 VolGroup00 -wi-ao 17.91G

LogVol01 VolGroup00 -wi-ao 1.97G

lv01 vg02 owi-ao 512.00M

lv01-shared vg02 swi--- 512.00M lv01 7.36

snap_2012081346132205130145 vg02 swi-a- 512.00M lv01

snap_2012081346132657130117 vg02 swi-ao 512.00M lv01

snap_2012081346132744130144 vg02 swi-ao 512.00M lv01

snap_2012081346133323130123 vg02 swi-a- 512.00M lv01

[root@localhost lv01]#

//删除快照测试

1.删除最老的快照 snap_2012081346132205130145，【在未重启原卷情况下，其他快照均正常】

[root@localhost lv01]# lvremove /dev/vg02/snap_2012081346132205130145

Do you really want to remove active logical volume snap_2012081346132205130145? [y/n]: y

Logical volume "snap_2012081346132205130145" successfully removed

[root@localhost lv01]# lvs

LV VG Attr LSize Origin Snap% Move Log Copy% Convert

LogVol00 VolGroup00 -wi-ao 17.91G

LogVol01 VolGroup00 -wi-ao 1.97G

lv01 vg02 owi-ao 512.00M

lv01-shared vg02 swi--- 512.00M lv01 0.18

snap_2012081346132657130117 vg02 swi-ao 512.00M lv01

snap_2012081346132744130144 vg02 swi-ao 512.00M lv01

snap_2012081346133323130123 vg02 swi-a- 512.00M lv01

[root@localhost lv01]# ll /mnt/snap_2012081346132657130117

总计 12

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost lv01]# ll /mnt/snap_2012081346132744130144

总计 12

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132730130130

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132731130131

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost lv01]# ll

总计 12

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132730130130

-rw-r--r-- 1 root root 0 08-28 13:45 2012081346132731130131

-rw-r--r-- 1 root root 0 08-28 13:46 2012081346132801130141

-rw-r--r-- 1 root root 0 08-28 13:46 2012081346132802130142

drwx------ 2 root root 12288 08-28 13:37 lost+found

[root@localhost lv01]#

//重启原卷，发现所有逻辑卷均失效，但可能因为前两个有挂载，再试一次

[root@localhost lv01]# fuser -m /dev/vg02/lv01

/dev/vg02/lv01: 3878c

[root@localhost lv01]# kill -9 3878

[root@localhost ~]# umount /dev/vg02/lv01

[root@localhost ~]# lvchange -an /dev/vg02/lv01

[root@localhost ~]# lvchange -ay /dev/vg02/lv01

/dev/mapper/vg02-snap_2012081346132657130117: read failed after 0 of 2048 at 0: 输入/输出错误

/dev/mapper/vg02-snap_2012081346132744130144: read failed after 0 of 2048 at 0: 输入/输出错误

[root@localhost ~]#

[root@localhost ~]# lvs

/dev/mapper/vg02-snap_2012081346132657130117: read failed after 0 of 2048 at 0: 输入/输出错误

/dev/mapper/vg02-snap_2012081346132744130144: read failed after 0 of 2048 at 0: 输入/输出错误

LV VG Attr LSize Origin Snap% Move Log Copy% Convert

LogVol00 VolGroup00 -wi-ao 17.91G

LogVol01 VolGroup00 -wi-ao 1.97G

lv01 vg02 owi-a- 512.00M

lv01-shared vg02 swi--- 512.00M lv010.18

snap_2012081346132657130117 vg02 swi-ao 512.00M lv01

snap_2012081346132744130144 vg02 swi-ao 512.00M lv01

snap_2012081346133323130123 vg02 swi--- 512.00M lv01

===========================================================================================================================================
可能的错误

lvcreate -s --sharedstore mikulas -c 64k -L 512M /dev/vg01/lv01

lvcreate:无法识别的选项"--sharedstore"

Error during parsing of command line

解答：重新给LVM打补丁并编译LVM工具，如上问3

[root@localhost LVM2. 2. 02. 73] # lvcreate -s --sharedstore mikulas -c 64k -L 512M /dev/vg02/lv01
WARNING : Unrecognised segment type multisnapshot
Cannot update volume group vg02 with unknown segments in it !

解答：打完LVM补丁后，为进行aclocal和autoconf对makefile进行整理

[root@localhost LVM2. 2. 02. 73] # lvcreate -s --sharedstore mikulas -L 512M -c 64K /dev/vg02/lv01
lvcreate : relocation error : lvcreate : symbol dm_tree_node_add_snapshot_shared_target, version Base not defined in file libdevmapper.so. 1. 02 with link time reference

解答：如果是64位的机器，需要将编译完的lvm同时不知道lib64下， 系统使用的是 /lib64/下的libdevmapper.so.1.02 ，那里并没被更新，从 /lib/拷过来就行了