RVC使用指南（一）-基本配置

RVC使用指南（一）-基本配置

https://mp.weixin.qq.com/s/yb27sFcLgddGj61lUVxhBg

 

vSAN 6.6 RVC系列主要讨论如何使用RVC管理vSAN环境。RVC是一个用来管理和自动化的命令行工具，本文所有实例均来自vSAN6.6。本文主要介绍基本配置所用到的命令及参数。

 

• vsan.enable_vsan_on_cluster

• vsan.disable_vsan_on_cluster

• vsan.cluster_change_autoclaim

• vsan.apply_license_to_cluster

• vsan.host_consume_disks

• vsan.cluster_set_default_policy

• vsan.host_wipe_vsan_disks

• vsan.host_wipe_non_vsan_disk

 

为了缩短命令，我对环境中的集群、虚拟机和ESXi主机使用了标记。这样可以在示例中使用~cluster、~vm和~esx来代替:

 

/localhost/DC> mark cluster ~/computers/VSAN-Cluster/
/localhost/DC> mark vm ~/vms/vma.virten.lab
/localhost/DC> mark esx ~/computers/VSAN-Cluster/hosts/esx1.virten.lab/

基本配置

vsan.enable_vsan_on_cluster [-d] ~cluster

在集群上启用vSAN，如果vSAN已经启用，可以使用命令启用或者禁用自动声明，如果未指定，默认开启自动声明。

 

• -d, --disable-storage-auto-claim: -禁用磁盘自动声明（磁盘的手动声明）

示例 1 - 在集群上启用vSAN

/localhost/DC> vsan.enable_vsan_on_cluster ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx1.virten.local: success
 esx2.virten.local: success
 esx3.virten.local: success

示例 2 -在启用了vSAN的集群上禁用自动声明

 

/localhost/DC> vsan.enable_vsan_on_cluster --disable-storage-auto-claim ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx1.virten.local: success
 esx2.virten.local: success
 esx3.virten.local: success

vsan.disable_vsan_on_cluster ~cluster

在集群上禁用vSAN。必须禁用VMware HA，才能禁用vSAN。此命令将使vsanDatastore（包括所有虚拟机）不可用，不会删除任何数据。

示例 1 - 在已启用HA的群集上禁用vSAN：

/localhost/DC> cluster.configure_ha --disabled ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx1.virten.local: success
 esx2.virten.local: success
 esx3.virten.local: success
/localhost/DC> vsan.disable_vsan_on_cluster ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx1.virten.local: success
 esx2.virten.local: success
 esx3.virten.local: success

vsan.cluster_change_autoclaim [-e|-d] ~cluster

此命令启用或禁用自动声明。启用自动声明后，Virtual SAN会自动使用所有未使用的磁盘。

 

• -e, --enable: Enable auto-claim

• -d, --disable: Disable auto-claim

示例 1 - 启用自动声明:

/localhost/DC> vsan.cluster_change_autoclaim --enable ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx2.virten.local: success
 esx3.virten.local: success
 esx1.virten.local: success

示例 2 - 禁用自动声明:

/localhost/DC> vsan.cluster_change_autoclaim --disable ~cluster
ReconfigureComputeResource VSAN-Cluster: success
 esx2.virten.local: success
 esx3.virten.local: success
 esx1.virten.local: success

vsan.apply_license_to_cluster [-k|-r] ~cluster
将vSAN许可应用于集群。此命令还会在所有主机上触发Null-Reconfigure，以确保启用自动声明后可以声明所有磁盘。

 

• -k, --license-key=: 被应用到集群的许可

• -r, --null-reconfigure, --no-null-reconfigure: 触发Null-Reconfigure以强制自动声明（默认值：true）

示例 1 - 应用vSAN许可并自动声明

/localhost/DC> vsan.apply_license_to_cluster -k 00000-00000-00000-00000-00000 ~cluster
VSAN-Cluster: Applying VSAN License on the cluster...
VSAN-Cluster: Null-Reconfigure to force auto-claim...
ReconfigureComputeResource VSAN-Cluster: success
 esx1.virten.local: success
 esx2.virten.local: success
 esx3.virten.local: success

示例 2 - 替换vSAN许可

/localhost/DC> vsan.apply_license_to_cluster -r -k 00000-00000-00000-00000-00000 ~cluster
VSAN-Cluster: Applying VSAN License on the cluster...

vsan.host_consume_disks [-f|-i] ~cluster|~host
使用主机上所有合格的磁盘。当主机具有多个SSD时，需要根据其型号指定适当的SSD。

 

• -f, --filter-ssd-by-model=: Regex to apply as SSD model filter

• -i, --filter-hdd-by-model=: Regex to apply as HDD model filter

示例 1 - 识别SSD并创建磁盘组。当有多个SSD时，这个会很方便：

/localhost/DC> esxcli.execute ~esx storage core device list
[...]
IsSSD: true
Model: "SanDisk SDSSDP06"
[...]
/localhost/DC> vsan.host_consume_disks ~esx --filter-ssd-by-model "SanDisk SDSSDP06"
 AddDisks esx1.virten.local: success

示例 2 -将所有合格的磁盘添加到新的或现有的磁盘组中：

/localhost/DC> vsan.host_consume_disks ~esx
 AddDisks esx1.virten.local: success

vsan.cluster_set_default_policy ~cluster ~policy
在集群上设置默认策略。当对象未定义VM存储策略时，默认使用该策略。策略定义对象在Virtual SAN中的存储方式

• hostFailuresToTolerate (Number of failures to tolerate)

• forceProvisioning (If VSAN can't fulfill the policy requirements for an object, it will still deploy it)

• stripeWidth (Number of disk stripes per object)

• cacheReservation (Flash read cache reservation)

• proportionalCapacity (Object space reservation)

默认策略: (("hostFailuresToTolerate" i1))

示例 1 - 更改默认策略以允许两个主机失败：

/localhost/DC> vsan.cluster_set_default_policy ~cluster '(("hostFailuresToTolerate" i2))'

示例 2 -确定现有主机上的默认策略并添加一条规则：

/localhost/DC> esxcli.execute ~esx vsan policy getdefault
+--------------+----------------------------------------------------------+
| Policy Class | Policy Value                                             |
+--------------+----------------------------------------------------------+
| cluster      | (("hostFailuresToTolerate" i1))                          |
| vdisk        | (("hostFailuresToTolerate" i1))                          |
| vmnamespace  | (("hostFailuresToTolerate" i1))                          |
| vmswap       | (("hostFailuresToTolerate" i1) ("forceProvisioning" i1)) |
+--------------+----------------------------------------------------------+
/localhost/DC> vsan.cluster_set_default_policy ~cluster '(("hostFailuresToTolerate" i1) ("stripeWidth" i2))'

vsan.host_wipe_vsan_disks [-d|-i|-a|-n|-f] ~host

删除主机上所有启用VSAN的磁盘的内容。如果在不使用force参数的情况下使用该命令，则会告诉用户要擦除的内容。

 

• -d, --disk=: Disk's canonical name, as identifier of disk to be wiped

• -i, --interactive: Select disks to wipe from given disk list, cannot be set together with parameter 'disks'

• -a, --allow-reduced-redundancy: Removes the need for disks worth of free space, by allowing reduced redundancy during disk wiping

• -n, --no-action: Take no action to protect data during disk wiping

• -f, --force: Forcely wipe disks without any confirmation

示例 1 - 在dry模式下运行命令：

/localhost/DC> vsan.host_wipe_vsan_disks ~esx
Would wipe disk WDC_WD1500HLFS (ATA WDC WD1500HLFS-0, ssd = false)
Would wipe disk WDC_WD1500HLFS (ATA WDC WD1500HLFS-0, ssd = false)
Would wipe disk WDC_WD1500HLFS (ATA WDC WD1500HLFS-0, ssd = false)
Would wipe disk SanDisk_SDSSDP064G (ATA SanDisk SDSSDP06, ssd = true)

NO ACTION WAS TAKEN. Use --force to actually wipe.
CAUTION: Wiping disks means all user data will be destroyed!

示例 2 - 擦除所有vSAN磁盘的内容:

/localhost/DC> vsan.host_wipe_vsan_disks ~esx --force
 RemoveDiskMapping esx1.virten.local: success

vsan.host_wipe_non_vsan_disk [-d|-f|-i] ~host

擦除vSAN分区以外的磁盘。此命令用于擦除以前用于其他文件系统(如VMFS、FAT、vFRC)的磁盘，现在将该磁盘用于虚拟SAN。当使用不带force参数的命令时，它只会告诉用户要擦除的内容。

 

• -d, --disk=: Disk to be wiped clean (multiple allowed)

• -f, --force: Do it for real

• -i, --interactive: Select disks to wipe from given disk list, cannot be set together with parameter 'disks'

示例 1 - 在dry模式下运行命令：

/localhost/DC> vsan.host_wipe_non_vsan_disk -d naa.6589cfc00000001917172ff67b90be59 ~esx
Gathering disk information for host vesx1.virten.lab
Done gathering disk information
Disks on host esx1.virten.lab:
Disk: FreeBSD iSCSI Disk (naa.6589cfc00000001917172ff67b90be59)
  Host: esx1.virten.lab
  Make/Model: FreeBSD iSCSI Disk
  Type: SSD
  Size: 50 GB
  Partition table:
Are you sure you want to delete all partitions? [y/N]
y
Would have deleted all partitions and all data, but
didn't. Run with --force to delete partitions and
erase all data on disk naa.6589cfc00000001917172ff67b90be59 for real.

示例 2 - 在交互模式下运行命令：

/localhost/DC>  vsan.host_wipe_non_vsan_disk ~esx/ -i
Gathering disk information for host vesx1.virten.lab
Done gathering disk information
Disks on host vesx1.virten.lab:
+-------+-----------------------------------------------------------------+-------------------------------------------------+
| Index | Disk info                                                       | Partition info                                  |
+-------+-----------------------------------------------------------------+-------------------------------------------------+
| 1     | Disk: FreeBSD iSCSI Disk (naa.6589cfc00000001917172ff67b90be59) | Partition table:                                |
|       | Make/Model: FreeBSD iSCSI Disk                                  |                                                 |
|       | Type: SSD                                                       |                                                 |
|       | Size: 50 GB                                                     |                                                 |
+-------+-----------------------------------------------------------------+-------------------------------------------------+
| 2     | Disk: FreeNAS iSCSI Disk (naa.6589cfc0000000572b71f35019e9c31f) | Partition table:                                |
|       | Make/Model: FreeNAS iSCSI Disk                                  |   1: 400.00 GB, type = vmfs ('flash-iscsi02')   |
|       | Type: SSD                                                       |                                                 |
|       | Size: 400 GB                                                    |                                                 |
+-------+-----------------------------------------------------------------+-------------------------------------------------+
Please input index of disks to be wiped, use comma to separate, input 'none' to skip:
1
Disk: FreeBSD iSCSI Disk (naa.6589cfc00000001917172ff67b90be59)
  Host: vesx1.virten.lab
  Make/Model: FreeBSD iSCSI Disk
  Type: SSD
  Size: 50 GB
  Partition table:
Are you sure you want to delete all partitions? [y/N]
y
Would have deleted all partitions and all data, but
didn't. Run with --force to delete partitions and

erase all data on disk naa.6589cfc00000001917172ff67b90be59 for real.

相关阅读：

ESXi主机上的NIC负载均衡：ESXCLI是首选

ESXi root 密码重置

如何在vSAN上安装ESXi？一条命令行就搞定！

ESXi 系统tmp目录100%，导致系统无响应，如何拯救？