vpxd service crashes on vCenter when performing a cross cluster vMotion

Products

• VMware vCenter Server

Issue/Introduction

/var/log/vmware/vmon/vmon.log on vCenter contains entries such as:

YYYY-MM-DDTHH:mm:ss Wa(03) host-xxxx <vpxd> Service exited unexpectedly. Crash count 0. Taking configured recovery action.
YYYY-MM-DDTHH:mm:ss Wa(03) host-xxxx <vpxd> Service exited unexpectedly. Crash count 1. Taking configured recovery action.

 

These can be correlated with /var/log/vmware/vpxd.log entries such as:

YYYY-MM-DDTHH:mm:ss warning vpxd[1436594] [Originator@6876 sub=sdrsLogger opID=xxxx] Datastore [vim.Datastore:datastore-xxx,DatastoreName] invalid size: capcaityBytes=2500000000000000, freeBytes=xxx

Environment

vCenter 7.0

vCenter 8.0

Cause

The capacity of the datastore is reported as 2,500,000,000,000,000 bytes in the above example, which corresponds to 2.5 Petabytes.

As per PlaceVm API does not work with datastores larger than 2PB when a datastore size is greater than 2PB, the PlaceVm API does not respond correctly. 

Resolution

VMware is aware of this issue and working to resolve this in a future release.

Workaround 1:
Users should use the datastores smaller than 2PB.

 

Workaround 2:

The vMotion can be completed by selecting a specific host and datastore manually rather than using DRS/Storage DRS for the placement.