上周高可用集群搭建完, 发现所有x86机器都没有搭建raid, 万一本地硬盘出问题机器可就挂了。后来安装raid重做系统。 下面是搭建lvm出错的地方：
[root@D2-04 ~]# pvcreate /dev/sdb1
Physical volume "/dev/sdb1" successfully created
解决办法如下:
dmsetup是一个底层的逻辑卷管理, dm应该是DeviceMapper的简称.
[root@D2-04 ~]# dmsetup status
mpathd: 0 2097152 multipath 2 0 0 0 1 1 A 01 0 8:64 A 0
mpathbp1: 0 306182772 linear
mpathc: 0 2097152 multipath 2 0 0 0 1 1 A 01 0 8:48 A 0
mpathcp1: 0 2096753 linear
mpathb: 0 306184192 multipath 2 0 0 0 1 1 A0 1 0 8:32 A 0
mpathdp1: 0 2096753 linear
[root@D2-04 ~]# dmsetup ls
mpathd (252:2)
mpathbp1 (252:5)
mpathc (252:3)
mpathcp1 (252:7)
mpathb (252:1)
mpathdp1 (252:6)
可以看到mpathd、mpathc、mpathb组成了multipath(多路径)
[root@D2-04 ~]# multipath -ll
mpathd (3600000e00d1000000010317200330000)dm-2 FUJITSU,ETERNUS_DX8000
size=1.0G features='0' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=130status=active
`-10:0:0:3 sde8:64 active ready running
mpathc (3600000e00d1000000010317200320000)dm-3 FUJITSU,ETERNUS_DX8000
size=1.0G features='0' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=130status=active
`-10:0:0:2 sdd8:48 active ready running
mpathb (3600000e00d10000000103172002b0000)dm-1 FUJITSU,ETERNUS_DX8000
size=146G features='0' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=130status=active
`-10:0:0:1 sdc8:32 active ready running
[root@D2-04 ~]# ll /dev/mapper/
total 0
crw-rw---- 1 root root 10, 58 Apr 27 15:40control
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathb -> ../dm-1
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathbp1 -> ../dm-5
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathc -> ../dm-3
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathcp1 -> ../dm-7
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathd -> ../dm-2
lrwxrwxrwx 1 root root 7 Apr 27 15:40 mpathdp1 -> ../dm-6
使用dmsetupremove xxx移除掉就可以创建PV了.
也可以使用dmsetupremove_all 移除所有.
编辑回复