Skip to main content
Solved

Check SSD Space Utilization of a vDisk/VM

  • Feb 17, 2020 12:03 PM UTC
  • 4 replies
  • 2058 views

Hi everybody,

I see a “flapping” warning message for one SSD on my 3-Node Hybrid-Cluster (2x 1,92TB SSD per Node).

Disk space usage for one or more disks on Controller VM ###.###.###.### is 88% which exceeds the threshold of 85%.

I checked curator runs which were successfull. 

From the I/O Metrics of a VM (Prism → VM → I/O Metrics) I only see the performance distribution (block size/read source usw.) for a point in time.

I know Nutanix DSF handles tiering but is there a command/way to figure out which vm/vdisk claims most SSD capacity?

Thanks for support,
Marc

Best answer by Alona

Marc, without understanding your environment it is difficult to point you in the right direction.

In addition, the disks you are referring to are for service data. Should the overall storage capacity be sufficient to accommodate the data then I would recommend opening a support case as the KB suggests.

View original
This topic has been closed for replies.

4 replies

Via Stargate:2009 i could figure out the working set size of a vDisk (2min/1h) which helps for now.
If you have any other ideas, please let me know.

Kind regards,

Marc

Alona
Nutanix Employee
Forum|alt.badge.img+5
  • AlonaNutanix Employee
  • Nutanix Employee
  • Feb 18, 2020 5:11 AM UTC

Hi Marc,

Have you seen this KB-3787 DiskSpaceUsageHigh?

Hi Alona,

thanks for your reply.

Yes, I read this article. The “flapping warning” is caused by the filled data/stargate-storage/disks/xxx.

So I end up investigating stargate:2009 which shows me the WSS of each vDisk.

Kind regards,

Marc

Alona
Nutanix Employee
Forum|alt.badge.img+5
  • AlonaNutanix Employee
  • Nutanix Employee
  • Answer
  • Feb 20, 2020 3:47 AM UTC

Marc, without understanding your environment it is difficult to point you in the right direction.

In addition, the disks you are referring to are for service data. Should the overall storage capacity be sufficient to accommodate the data then I would recommend opening a support case as the KB suggests.

Terms & ConditionsAccessibility statement