Applicable Products
- QuTS hero h6.0.0 or later
- Storage Manager
Scenario
I have created a VJBOD storage pool on a local NAS that uses VJBOD disks created from LUNs on a remote high-availability (HA) cluster. When a failover occurs in the remote HA cluster, the VJBOD storage pool status on my local NAS changes to "Error", and all ongoing I/O operations are interrupted.
Root Cause
Although the VJBOD service switches to the other node during an HA failover, this process is still registered as a connection loss. This causes the operating system on the local NAS to detect I/O errors and change the VJBOD storage pool status to "Error" as a data protection measure. This is currently the expected behavior.
Solution
You can manually recover the VJBOD storage pool without deleting or rebuilding it by following these steps:
- Verify the remote HA cluster has recovered from the failover.
- Log in to the HA cluster as an administrator.
- Open High Availability Manager.
- Go to Cluster.
- Verify the nodes have switched roles.
- Verify the cluster status is "Good".
- Recover the VJBOD storage pool on the local NAS.
- Log in to the local NAS as an administrator.
- Open Storage Manager.
- Go to Storage Space.
- Identify the VJBOD storage pool with the "Error" status.
- Under Action, click ⋮ and select Manage.
The pool management window opens. - Go to the RAID tab.
- Click Manage and select Recover Storage Pool.
The system recovers the VJBOD storage pool. - Once the recovery is complete, the VJBOD storage pool status changes to "Ready".
You can now access the data in the pool again.
Important
Because VJBOD does not currently support HA failovers, you would need to manually recover your VJBOD storage pool every time a failover occurs on the remote HA cluster. Before you continue using this setup, please evaluate whether this limitation meets your production requirements. You can consider other methods of storage expansion or use LUNs from a standalone NAS instead of a NAS in an HA cluster.
Further Reading
适用产品
- QuTS hero h6.0.0 或更高版本
- 存储管理器
场景
我在本地 NAS 上创建了一个 VJBOD 存储池，该池使用从远程高可用性 (HA) 群集的 LUN 创建的 VJBOD 磁盘。当远程 HA 群集中发生故障转移时，我的本地 NAS 上的 VJBOD 存储池状态变为“错误”，所有正在进行的 I/O 操作被中断。
根本原因
尽管在 HA 故障转移期间 VJBOD 服务会切换到另一个节点，但此过程仍被注册为连接丢失。这导致本地 NAS 上的操作系统检测到 I/O 错误，并将 VJBOD 存储池状态更改为“错误”作为数据保护措施。这是当前的预期行为。
解决方案
您可以通过以下步骤手动恢复 VJBOD 存储池，而无需删除或重建它：
- 验证远程 HA 群集已从故障转移中恢复。
- 以管理员身份登录到 HA 群集。
- 打开 High Availability Manager。
- 前往 群集。
- 验证节点已切换角色。
- 验证群集状态为“良好”。
- 在本地 NAS 上恢复 VJBOD 存储池。
- 以管理员身份登录本地 NAS。
- 打开存储管理器。
- 前往存储空间。
- 识别状态为“错误”的 VJBOD 存储池。
- 在操作下，点击⋮ 并选择 管理。
池管理窗口打开。 - 前往RAID选项卡。
- 点击管理并选择 恢复存储池。
系统将恢复 VJBOD 存储池。 - 恢复完成后，VJBOD 存储池状态变为“就绪”。
您现在可以再次访问池中的数据。
重要提示
由于 VJBOD 目前不支持 HA 故障转移，每次远程 HA 群集发生故障转移时，您需要手动恢复 VJBOD 存储池。在继续使用此设置之前，请评估此限制是否符合您的生产需求。您可以考虑其他存储扩展方法，或使用独立 NAS 的 LUN，而不是 HA 群集中的 NAS。
进一步阅读
