/dev/sdxx is apparently in use by the system; will not make a filesystem here! 解决方法


 在存储上共享了一个500G的空间,映射到Linux系统提供上,环境由2个节点组成。

 

一. 测试一: 直接mount

用fdisk 格式化之后如下:

[root@rac1 u01]# fdisk -l

……

Disk /dev/sdk: 536.8 GB, 536870912000 bytes

255 heads, 63 sectors/track, 65270cylinders

Units = cylinders of 16065 * 512 = 8225280bytes

 

  Device Boot      Start         End      Blocks  Id  System

/dev/sdk1               1       65270  524281243+  83  Linux

……

 [root@rac1u01]#

 

但是创建文件系统时报错:

[root@rac1 u01]# mkfs -t ext3 /dev/sdk1

mke2fs 1.39 (29-May-2006)

/dev/sdk1 isapparently in use by the system; will not make a filesystem here!

 

提示/dev/sdk1 正在被使用。 /dev/sdk1 正在被DM管理,所以我们创建文件系统时提示报错,我们手工的移除,就可以正常的创建文件系统,操作如下:

 

[root@rac1 u01]# dmsetup status

mpath2: 0 2097152 multipath 2 0 1 0 1 1 A 01 0 8:16 A 0

mpath11p1: 0 1048562487 linear

mpath9: 0 209715200 multipath 2 0 1 0 1 1 A0 1 0 8:128 A 0

mpath8: 0 629145600 multipath 2 0 1 0 1 1 A0 1 0 8:112 A 0

mpath7: 0 629145600 multipath 2 0 1 0 1 1 A0 1 0 8:96 A 0

mpath6: 0 2097152 multipath 2 0 1 0 1 1 A 01 0 8:80 A 0

mpath5: 0 2097152 multipath 2 0 1 0 1 1 A 01 0 8:64 A 0

mpath11: 0 1048576000 multipath 2 0 1 0 1 1A 0 1 0 8:160 A 0

mpath4: 0 2097152 multipath 2 0 1 0 1 1 A 01 0 8:48 A 0

mpath10: 0 209715200 multipath 2 0 1 0 1 1A 0 1 0 8:144 A 0

mpath3: 0 2097152 multipath 2 0 1 0 1 1 A 01 0 8:32 A 0

 

[root@rac1 u01]# dmsetup remove_all

 

[root@rac1 u01]# dmsetupstatus         

No devices found

 

[root@rac1 u01]# mkfs -text3 /dev/sdk1 

mke2fs 1.39 (29-May-2006)

Filesystem label=

OS type: Linux

Block size=4096 (log=2)

Fragment size=4096 (log=2)

65536000 inodes, 131070310 blocks

6553515 blocks (5.00%) reserved for thesuper user

First data block=0

Maximum filesystem blocks=4294967296

4000 block groups

32768 blocks per group, 32768 fragments pergroup

16384 inodes per group

Superblock backups stored on blocks:

       32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,

       4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,

       102400000

 

Writing inode tables: done                           

Creating journal (32768 blocks): done

Writing superblocks and filesystemaccounting information: done

 

This filesystem will be automaticallychecked every 36 mounts or

180 days, whichever comes first.  Use tune2fs -c or -i to override.

 

--创建文件系统成功。

 

--mount 成功:

[root@rac1 u01]# mount /dev/sdk1/u01/backup

[root@rac1 u01]# df -lh

Filesystem            Size  Used Avail Use% Mounted on

/dev/sda3              59G   22G  35G  39% /

/dev/sda1             996M   51M 894M   6% /boot

tmpfs                  32G     0  32G   0% /dev/shm

/dev/sda4             145G  188M 138G   1% /u01/dave

/dev/sdk1             493G  198M 467G   1% /u01/backup

 

 

--修改/etc/fstab 文件,让开机自动挂载:

[root@rac2 mapper]# vi /etc/fstab

 

LABEL=/                 /                       ext3    defaults        1 1

LABEL=/boot             /boot                   ext3    defaults        1 2

tmpfs                   /dev/shm                tmpfs   defaults        0 0

devpts                 /dev/pts                devpts  gid=5,mode=620  0 0

sysfs                   /sys                    sysfs   defaults        0 0

proc                    /proc                   proc    defaults        0 0

LABEL=SWAP-sda2         swap                    swap   defaults        0 0

/dev/sdk1               /u01/backup             ext3    defaults        0 0

 

 

但重启后测试,该文件不能正常挂载,手工挂载也会失败。

 

所以,这种解决方法不行。

 

 

补充内容:

Device mapper 是 Linux 2.6 内核中提供的一种从逻辑设备到物理设备的映射框架机制,在该机制下,用户可以很方便的根据自己的需要制定实现存储资源的管理策略,如条带化,镜像,快照等. 当前比较流行的 Linux 下的逻辑卷管理器如 LVM2(Linux Volume Manager 2 version)、EVMS(EnterpriseVolume Management System)、dmraid(Device Mapper RaidTool)等都是基于该机制实现的. 只要用户在用户空间制定好映射策略,按照自己的需要编写处理具体IO请求的 target driver插件,就可以很方便的实现这些特性.

 

Device Mapper主要包含内核空间的映射和用户空间的device mapper库及dmsetup工具.

 

 

二. 实验二: 使用Multipath

 

关于Multipath的配置说明,参考:

Multipath实现LUN设备名称的持久化

http://blog.csdn.net/tianlesoftware/article/details/5979061

 

--获取wwid

[root@rac1 mapper]# /sbin/scsi_id -g -u -s/block/sdk

3690b11c00022bc0e000003e55105b786

 

--修改multipath.conf 文件:

[root@rac1 mapper]# vi /etc/multipath.conf

 

multipaths {

        multipath {

                wwid                   3690b11c00022bc0e000003e55105b786

                alias                   backup

                path_grouping_policy    multibus

                path_checker            readsector0

                path_selector           "round-robin 0"

                failback                manual

                rr_weight               priorities

                no_path_retry           5

       }

#      multipath {

#               wwid                    1DEC_____321816758474

#               alias                   red

#      }

}

 

"/etc/multipath.conf" 177L, 4832Cwritten

 

--重启multipath

[root@rac1 mapper]# service multipathdrestart

Stopping multipathd daemon:                                [  OK  ]

Starting multipathd daemon:                                [  OK  ]

 

--检查文件:

[root@rac1 mapper]# cd /dev/mapper/

[root@rac1 mapper]# ll

total 0

brw-rw---- 1 root disk253,  9 Feb 20 12:35 backup

brw-rw---- 1 root disk253, 10 Feb 20 12:35 backupp1

crw------- 1 root root  10, 60 Feb 20 12:35 control

brw-rw---- 1 root disk 253,  8 Feb 20 12:35 mpath10

brw-rw---- 1 root disk 253,  0 Feb 20 12:35 mpath2

brw-rw---- 1 root disk 253,  1 Feb 20 12:35 mpath3

brw-rw---- 1 root disk 253,  2 Feb 20 12:35 mpath4

brw-rw---- 1 root disk 253,  3 Feb 20 12:35 mpath5

brw-rw---- 1 root disk 253,  4 Feb 20 12:35 mpath6

brw-rw---- 1 root disk 253,  5 Feb 20 12:35 mpath7

brw-rw---- 1 root disk 253,  6 Feb 20 12:35 mpath8

brw-rw---- 1 root disk 253,  7 Feb 20 12:35 mpath9

 

--mount 文件:

[root@rac1 mapper]# mount/dev/mapper/backupp1 /u01/backup

 

--检查mount

[root@rac1 mapper]# df -lh

Filesystem            Size  Used Avail Use% Mounted on

/dev/sda3              59G   22G  34G  39% /

/dev/sda1             996M   51M 894M   6% /boot

tmpfs                  32G  364M  32G   2% /dev/shm

/dev/sda4             145G  188M 138G   1% /u01/dave

/dev/mapper/backupp1  493G 198M  467G   1% /u01/backup

 

 

修改/etc/fstab文件后,重启可以正常的自动挂载。但是这里是2个节点,存储配置的也是共享的。 但是在节点1创建的文件,在节点2上不能识别,经测试,只有在重新mount之后,才可以看到另一个节点创建的文件。

 

 

测试步骤如下:

[root@rac1 backup]# ll

total 24

-rw-r--r-- 1 root root     0 Feb 20 12:57 bl

drwxr-xr-x 2 root root  4096 Feb 20 12:55 dave

-rw-r--r-- 1 root root     5 Feb 20 12:55 dvd

drwx------ 2 root root 16384 Feb 20 12:10lost+found

 

--创建文件orcl

[root@rac1 backup]# touch orcl

 

--在节点2 umount 目录:

[root@rac2 backup]# umount /u01/backup

umount: /u01/backup: device is busy

umount: /u01/backup: device is busy

[root@rac2 backup]# fuser -km /u01/backup

/u01/backup:          9848c

 

[root@rac2 ~]# df -lh

Filesystem            Size  Used Avail Use% Mounted on

/dev/sda3              70G   20G  46G  31% /

/dev/sda1             996M   51M 894M   6% /boot

tmpfs                  32G  364M  32G   2% /dev/shm

/dev/mapper/backupp1  493G 198M  467G   1% /u01/backup

[root@rac2 ~]# umount  /u01/backup

 

--确认umount 成功:

[root@rac2 ~]# df -lh

Filesystem            Size  Used Avail Use% Mounted on

/dev/sda3              70G   20G  46G  31% /

/dev/sda1            996M   51M 894M   6% /boot

tmpfs                  32G  364M  32G   2% /dev/shm

 

--再次mount

[root@rac2 ~]# mount /dev/mapper/backupp1 /u01/backup

[root@rac2 ~]# cd /u01/backup

[root@rac2 backup]# ll

total 24

-rw-r--r-- 1 root root     0 Feb 20 12:57 bl

drwxr-xr-x 2 root root  4096 Feb 20 12:55 dave

-rw-r--r-- 1 root root     5 Feb 20 12:55 dvd

drwx------ 2 root root 16384 Feb 20 12:10lost+found

-rw-r--r-- 1 root root     0 Feb 20 14:34 orcl

[root@rac2 backup]#

 

这次就在节点2上看到我们节点1上创建的文件了。

 

 

 

 

---------------------------------------------------------------------------------------

版权所有,文章允许转载,但必须以链接方式注明源地址,否则追究法律责任!

Skype:    tianlesoftware

QQ:       [email protected]

Email:    [email protected]

Blog:     http://blog.csdn.net/tianlesoftware

Weibo:    http://weibo.com/tianlesoftware

Twitter:  http://twitter.com/tianlesoftware

Facebook: http://www.facebook.com/tianlesoftware

Linkedin: http://cn.linkedin.com/in/tianlesoftware


你可能感兴趣的:(/dev/sdxx is apparently in use by the system; will not make a filesystem here! 解决方法)