How to Handle the Datastore Exception Caused by LUN Removal after Datastore Capacity Expansion? Highlighted

Latest reply: Aug 13, 2019 19:25:44 185 2 5 1

Issue Description】

A customer use LUNs to expand the capacity of data storage CNA04-1 on the FC portal. After the capacity expansion is complete, the mapping between the LUNs used for capacity expansion and the host is removed from the storage and mapped to other hosts for data storage. As a result, the data storage is abnormal.


【Version】 FusionCompute XEN version


Handling Process】


1. According to the task tracing result, we can confirm that the capacity of the data store is increased.

2. On the summary tab page of the storage pool, find the number of storage devices of the data is 2, where *03LUN is the original LUN, *09LUN is the LUN used for capacity expansion, and *09LUN has been removed from the host.

153215yixiv6f8ezxe9mmv.png?image.png

 

【Root Cause】


When data storage capacity is expanded, the VIMS file system will write VIMS metadata to the expanded LUN. After the LUN*09 is unmapped from the storage, the host cannot detect the LUN*09 and the metadata integrity of the data is damaged. As a result, the data storage is abnormal.


【Solution】

              

Note: In practice, the ID of the LUN shall prevail.

1. Remap the *09LUN used for capacity expansion to all hosts in the original environment.


2. After the association is successful, scan the storage device on the FC portal. After the storage device is successfully scanned, the data storage status can be restored to the associated state.


3. After the user unmaps the LUN, the data storage is added to hosts in other environment. As a result, the VIMS metadata of the data storage CNA04-1 is incomplete. In scenarios where only LUNs are demapped, the metadata of data storage will not be affected. If the metadata of the data storage is not damaged, then the restoration is not necessary..


4. Run the o2image command to check whether the metadata of the data storage is damaged. Note that o2image is used to test the source LUN of the data storage instead of the LUN used for capacity expansion. If the result is not successful and an error is reported, indicating that the metadata of the data storage is damaged and it needs to be repaired. If the test is successful, indicating the data storage has been restored.


153256pnwkl8hcf9jz9ew6.png?image.png


5. Determine whether running VMs exists on the data storage, and if yes, disable all VMs on the data storage first.


6. Refer to the section "VIMS File System Damage of Shared Storage" in the product documentation of the corresponding version to repair the data storage. After the repair is successful, the data storage can be restored to normal. 


  • x
  • convention:

ganmianpi
Admin Created Aug 13, 2019 03:42:29 Helpful(1) Helpful(1)

How to Handle the Datastore Exception Caused by LUN Removal after Datastore Capacity Expansion?-3032001-1
  • x
  • convention:

wissal
MVE Created Aug 13, 2019 19:25:44 Helpful(0) Helpful(0)

Thanks for sharing this solution
  • x
  • convention:

Telecommunications%20engineer%2C%20currently%20senior%20project%20manager%20at%20an%20operator%2C%20partner%20of%20Huawei%2C%20in%20the%20radio%20access%20network%20department%2C%20for%2020%20years%20I%20managed%20several%20types%20of%20projects%2C%20for%20the%20different%20nodes%20of%20the%20network.
Login and enjoy all the member benefits

Login and enjoy all the member benefits

Login