Solved

NCC Health Check: Garbage egroups

  • 1 February 2021
  • 1 reply
  • 174 views

Userlevel 1
Badge +9

I’m seeing:

Detailed information for garbage_egroups_size_check:

Node 10.1.11.32:

FAIL: 67% of disk 415104829 occupied by garbage egroups

Refer to KB 1574 (http://portal.nutanix.com/kb/1574) for details on garbage_egroups_size_check or Recheck with: ncc health_checks stargate_checks garbage_egroups_size_check

on 2019.02.11 LTS

Disk 415104829 (SSD) was added yesterday. I also added one ssd to 2 nodes tonight. The new ssd’s are very busy right now.

icon

Best answer by Chaitrali Deshpande 1 February 2021, 17:48

View original

This topic has been closed for comments

1 reply

Userlevel 2
Badge +4

Hello @kfr- 

Garbage egroups is data that is no longer in use by the system but that is still taking up space. Due to the way the Nutanix file system works, it is expected to have a small amount of garbage on the cluster. However, the system is periodically cleaning up garbage egroups in the background via Curator scans.

Things that need to be check:

  1. Make sure you are on the latest NCC version (4.0.1), as the thresholds for this check were adjusted in NCC 3.9.0 and above for improved accuracy.
  2. Make sure you are on the 5.10+ AOS version.
  3. Make sure the curator service is running. You can check this using two commands:
    • nutanix@cvm$ cluster status
    • nutanix@cvm$watch -d genesis status
  4. Garbage details are found using nutanix@cvm$curator_cli display_garbage_report this command.

Please let me know if you have further questions.