Question

Reclaim storage space after VMs moved to different storage containers

  • 18 November 2022
  • 3 replies
  • 670 views

Userlevel 1
Badge +2

Hello everybody, I started using Storage VM migration …

 

vm.update_container VMNAME container=test-container

 

… to move VMs from an old to a new container. Big VMs have been moved.

 

I would like to delete the old storage containers, but space usage is still high, allthough VM have been moved to another storage container.

 

I am aware of the directories …

 

.acropolis, .file_repo and .ngt

 

… and those directories are empty now.

 

How can I reclaim space in a storage container, when VMs have been moved, the Recycle Bin has been emptied.

No VMs have been deleted, only moved, so the 36 hours rule for space reclaimation doesn’t seem to apply here?

Any thoughts?

Regards,

Didi7

 


This topic has been closed for comments

3 replies

Userlevel 5
Badge +6

I know you didn't delete them, I just put the full conditions. In this case you have to contact Nutanix Support.

Userlevel 1
Badge +2

Hello Moustafa,

yes, that’s what I read about deleting VMs, but I did not delete VMs, I just moved them from one to another container, more than 24 hours have past, but not a single byte has been reclaimed so far.

I will let pass the upcoming weekend and check on monday again.

Remember I have not deleted VMs, I just have fully moved them to another container and the old container did not drop in space usage so far.

Regards,

Didi7

Userlevel 5
Badge +6

Hello @Didi7 

There are 2 main reasons for delay in reclaiming space:

 The service responsible for cleaning up space, Curator, takes time to perform scans on the cluster and spreads the storage across the disks and carefully removes the correct files (egroups) while maintaining redundancy for any data you want to keep.

It takes a few Curator scans before all expected free space is available again. To remove more data, it will take longer time for Curator to clean up data relative to the existing workload of the cluster.

Note: It is always recommended to keep your cluster version up-to-date, as there are significant improvements to Curator and its scan efficiency on the latest releases for faster clean up operations. 

AOS version 5.18 and above have Recycle Bin feature enabled on the cluster by default.

When the Recycle Bin is enabled, AOS creates a Recycle Bin associated with the storage container configured in your cluster. If you then delete a storage entity (like a guest VM or volume group vDisk), the Recycle Bin retains vDisk and configuration data files for up to 24 hours.

In case after 36 hours, space is not reclaimed and does not appear as free space, consider engaging Nutanix Support.