VxRail: Increase Repo Space threshold values in VCF 5.2 onwards.

Products

• VMware Cloud Foundation

Issue/Introduction

This article exists to guide users about the minimum repo disk space required for vxrail upgrade in VCF 5.2 onward. Since the necessary steps are not available via the User Interface (UI) or API, this article serves as an essential resource for ensuring a smooth upgrade process.

Symptoms:

The LCM bundle folder '${nfs_mount_dir}' does not have enough free disk space.
As we are adding functionality for unified bundle. Bundle size will be increased. So we need more empty space to have both zip and extracted file in sddc-manager.

Changes Required:

• Current threshold values:
  "sddc_manager_controller_vm_nfs_repo_disk_space_low_threshold_in_gb": 50,
"sddc_manager_controller_vm_nfs_repo_disk_space_high_threshold_in_gb": 110,
  
  
• New threshold values:
  "sddc_manager_controller_vm_nfs_repo_disk_space_low_threshold_in_gb": 70,
"sddc_manager_controller_vm_nfs_repo_disk_space_high_threshold_in_gb": 130,

Environment

VMware Cloud foundation 5.x

Cause

From VCF 5.2 onwards, VXRail upgrade will use a unified bundle, which will be double the size of the current bundle as it will contain both minor and major version bundles in one zip. Also, this bundle will be extracted by sddc manager. So more empty disk space is needed.

Resolution

Currently there is no resolution.

Workaround:

When there is a warning/error on repo disk space, the user can do the disk space cleanup by manually running bundle_cleanup.py script for some old bundles (which is not required).
Follow KB 312173 for bundle cleanup.