Virtual machine hung\unresponsive and not restarted by vSphere HA on other host.
search cancel

Virtual machine hung\unresponsive and not restarted by vSphere HA on other host.

book

Article ID: 403423

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

  • Virtual machines are unresponsive saved on one datastore.
  • Web console shows blank screen of VM.
  • Esxi host may go unresponsive if management agents are restarted when troubleshooting the issue.
  • vmkernel logs reports below logs
    YYYY-MM-DDTHH:MM:SS  In(182) vmkernel: cpu75:2097338)ScsiDeviceIO: 4672: Cmd(0x45ea24084740) 0x2a, CmdSN 0x145fc90 from world 7012903 to dev "naa.<id>" failed H:0x0 D:0x2 P:0x8 Valid sense data: 0x7 0x27 0x7
    YYYY-MM-DDTHH:MM:SS  In(182) vmkernel: cpu41:2097307)ScsiDeviceIO: 4672: Cmd(0x45ba4559e180) 0x2a, CmdSN 0x80000011 from world 7012903 to dev "naa.<id>" failed H:0x0 D:0x2 P:0x8 Valid sense data: 0x7 0x27 0x7

    This storage command H:0x0 D:0x2 P:0x8 Valid sense data: 0x7 0x27 0x7 reports that underlying storage is full and could not allocate more space.

  • Identify the datastore with naa.<id> is datastore where VMs are saved. Datasstore name can be looked by running command localcli storage vmfs extent list
  • vSphere HA will not restart VM on other datastore.
  • If guest Monitoring is enabled, HA might try to restart VM however, due to datastore not allowing any IO operations it will fail to restart.

Environment

VMware vSphere Esxi 7.0

VMware vSphere Esxi 8.0

Vmware vCenter server 7.0

Vmware vCenter server 8.0

Cause

This issue occurs when backing storage is fully utilized and there is no more space to be allocated.

vSphere HA will not be able to restart VM on another host using the same full datastore. 

 

Resolution

To resolve this issue, increase the space on datastore.

We can also storage vmotion the VM to free up space from datastore.

or Delete few files from datastore which are not required.

 

Additional Information

Powered by Wolken Software