ZFS - 存储池管理

ZFS是通过存储池来管理ZFS文件系统的，那么，我们首先因该知道如何管理存储池。
我们先从存储池操作相关的命令开始。

虚拟设备类型

1. 磁盘

root@zfsubuntu:~# zpool status disk_pool
pool: disk_pool
state: ONLINE
scan: none requested
config:

NAME        STATE     READ WRITE CKSUM
disk_pool   ONLINE       0     0     0
sdb       ONLINE       0     0     0

errors: No known data errors

2. 分区

root@zfsubuntu:~# zpool status par_pool
  pool: par_pool
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    par_pool    ONLINE       0     0     0
      sdb1      ONLINE       0     0     0

errors: No known data errors

3. 文件

root@zfsubuntu:~# zpool status file_pool
  pool: file_pool
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    file_pool               ONLINE       0     0     0
      /root/forzfsdiskfile  ONLINE       0     0     0

errors: No known data errors

存储池的创建与删除

1. 创建存储池

创建成功之后，存储池的顶层文件系统的默认挂载点是/pool_name. 当然，可以使用-m选项修改默认挂载点。

a) 创建普通的存储池

root@zfsubuntu:~# zpool create tank /dev/sdb
root@zfsubuntu:~# zpool status tank
  pool: tank
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    tank        ONLINE       0     0     0
      sdb       ONLINE       0     0     0

errors: No known data errors

b) 创建镜像存储池

root@zfsubuntu:~# zpool create tank mirror /dev/sdb /dev/sdc
root@zfsubuntu:~# zpool status tank
  pool: tank
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    tank        ONLINE       0     0     0
      mirror-0  ONLINE       0     0     0
        sdb     ONLINE       0     0     0
        sdc     ONLINE       0     0     0

errors: No known data errors

c) 创建带有日志设备的存储池

root@zfsubuntu:~# zpool create tank /root/forzfsdiskfile1 log mirror /dev/sdb /dev/sdc
root@zfsubuntu:~# zpool status tank
  pool: tank
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    tank                     ONLINE       0     0     0
      /root/forzfsdiskfile1  ONLINE       0     0     0
    logs
      mirror-1               ONLINE       0     0     0
        sdb                  ONLINE       0     0     0
        sdc                  ONLINE       0     0     0

errors: No known data errors

d) 创建带有高速缓存的存储池

root@zfsubuntu:~# zpool create tank /root/forzfsdiskfile1 cache  /dev/sdb /dev/sdc
root@zfsubuntu:~# zpool status tank
  pool: tank
 state: ONLINE
  scan: none requested
config:

    NAME STATE     READ WRITE CKSUM
    tank                     ONLINE       0     0     0
      /root/forzfsdiskfile1  ONLINE       0     0     0
    cache
      sdb                    ONLINE       0     0     0
      sdc                    ONLINE       0     0     0

errors: No known data errors

2. 删除（销毁）存储池

root@zfsubuntu:~# zpool destroy tank
root@zfsubuntu:~#

虚拟设备管理

1. 添加虚拟设备

root@zfsubuntu:~# zpool status tank
pool: tank
state: ONLINE
scan: none requested
config:

NAME                     STATE     READ WRITE CKSUM
tank                     ONLINE       0     0     0
/root/forzfsdiskfile1  ONLINE       0     0     0

errors: No known data errors
root@zfsubuntu:~# zpool add -f tank mirror /dev/sdb /dev/sdc
root@zfsubuntu:~# zpool status tank
pool: tank
state: ONLINE
scan: none requested
config:

NAME                     STATE     READ WRITE CKSUM
tank                     ONLINE       0     0     0
/root/forzfsdiskfile1  ONLINE       0     0     0
mirror-1               ONLINE       0     0     0
sdb                  ONLINE       0     0     0
sdc                  ONLINE       0     0     0

errors: No known data errors

2. 移除虚拟设备

root@zfsubuntu:~# zpool remove tank mirror-1
cannot remove mirror-1: only inactive hot spares, cache, top-level, or log devices can be removed

从这里我们看出，只有非活动状态的热备份设备，高速缓存设备，top-level级别的设备以及日志设备可以被移除。
至于上面给出的例子，mirror-1确实是top-level设备，但是移除失败，应该是一个bug。（这里的测试环境是Ubuntu，即zfsonlinux。在solaris环境下，执行OK）

至于对zpool的管理，有详细的说明，就不再过多说明了，后面，我们主要来看看这些操作时如何实现的。