RDM disks are inaccessible following Microsoft Cluster failover

Products

• VMware vSphere ESXi

Issue/Introduction

On Microsoft Cluster node failover

• vCenter reports "Some of the disks of the virtual machine #### failed to load. The information present to them in the virtual machine configuration may be incomplete"
• Microsoft Failover Cluster Manager reports "Ownership of cluster disk ## has been unexpectedly lost by his node"

 

Environment

VMware vSphere ESXi 7.x
VMware vSphere ESXi 8.x

 

Cause

APD events seen on host where failover VM resides.

Logs report All Path Down events to RDM luns mapped to vmdks.

/var/log/vmkernel.log

vmkernel: cpu56:2097696)ScsiDevice: 5738: Device state of naa.#### set to APD_START; token num:1
vmkernel: cpu56:2097696)StorageApdHandler: 1191: APD start for 0x4309ed43eec0 [naa.####]
vmkwarning: cpu64:2097923)WARNING: NMP: nmpDeviceAttemptFailover:718: Retry world failover device "naa.####" - failed to issue command due to Not found (APD)

 

Resolution

Engage Storage Vendor to check connectivity to storage.