ESXi Hosts Become Unresponsive Due to Stale NFS 4.1 Locks on NetApp QTree-Based Datastores

Products

• VMware vCenter Server

Issue/Introduction

Multiple ESXi hosts within a cluster become unresponsive and enter a "Not Responding" state. The issue is observed on NFS 4.1 datastores hosted on NetApp storage, where hosts repeatedly attempt to access datastore metadata files and report file access failures.

Affected environments may experience the following symptoms:

• ESXi hosts become unresponsive and may require management intervention.
• Virtual machines running on affected hosts become inaccessible.
• Host resources become exhausted due to repeated storage access retries.
• Rebooting affected hosts does not resolve the issue.
• vmkernel logs contain repeated NFS 4.1 file open failures reporting Busy errors.
  

  2026-05-24T05:29:27.611Z In(182) vmkernel: cpu194:2101402 opID=85796501)NFS41: NFS41FileOpenFile:3008: Open of 0x43112126c540 name xxxxx  with openFlags 0x101 failed: Busy
  2026-05-24T05:29:27.611Z Wa(180) vmkwarning: cpu194:2101402 opID=85796501)WARNING: NFS41: NFS41FileOpOpenFile:3743: Open of obj 0x43112126c540 name xxxxx  (fhID 0x308588cd) failed: Busy 
  2026-05-24T05:29:27.611Z In(182) vmkernel: cpu193:2101410 opID=123351cS)NFS41: NFS41File0penFile:3008: Open of 0x43112126c540 name xxxxx with openFlags 0xl01 failed: Busy
  2026-05-24T05:29:27.861Z In(182) vmkernel: cpu193:2101410 opID=123351cSJNFS41: NFS41FileOpenFile:3008: Open of 0x43112126c540 name shard_xxxxx t with openFlags 0x101 failed: Busy 
  2026-05-24T05:29:28.362Z In(182) vmkernel: cpu193:2101410 oplD=123351cS)NFS41: NFS41FileOpenFile:3008: Open of 0x43112126c540 name shard_xxxxx  with openflags 0x101 failed: Busy 
  2026-05-24T05:29:29.362Z ln(182) vmkernel: cpu193:2101410 opl0=123351c5)NFS41: NFS41FileOpenFile:3008: Open of 0x43112126c540 name shard_xxxxx  with openFlags 0x101 failed: Busy 
  
  

Environment

VMware Esxi 8.x

Cause

The issue occurs when stale NFS 4.1 session and lock information remains on the storage server.

In this condition:

• The NFS server retains stale lock ownership information.
• ESXi hosts continuously retry access to locked metadata files.
• File locks are not properly released following a storage or session disruption.
• The storage server may experience session exhaustion due to repeated retry attempts.

As a result, ESXi hosts are unable to access required datastore files, causing increasing resource consumption and eventually leading to host unresponsiveness.

The issue is commonly observed in environments where VMware NFS datastores are hosted on NetApp QTree-based storage configurations.

Resolution

NetApp recommends using FlexVol volumes for VMware NFS datastores and generally does not recommend QTree-based configurations for VMware integrations. Refer the below Netapp article.
https://docs.netapp.com/us-en/ontap-apps-dbs/vmware/vmware-vsphere-datastores-top.html#selecting-a-storage-protocol  

Work-around:
Perform LIF-migration from the storage end which should terminate stale client sessions and release orphaned NFS 4.1 file locks.