All datastores on an ESXi host show 0 B capacity after a physical RAID drive replacement.

Products

• VMware vSphere ESXi

Issue/Introduction

Following a physical RAID drive replacement on an ESXi host, all local datastores on that host report 0 bytes in size. All virtual machines on the affected ESXi host also become inaccessible.

Environment

VMware vSphere ESXi 7.0.x

VMware vSphere ESXi 8.0.x

Cause

A RAID rebuild takes place, but its status does not update on the ESXi host.

1. Login to the host using SSH and go to var/run/log and check if you can get the vmkwarning/vobd or vmkernel log. However, an Input/output error occurs if the log/scratch location is the local datastore.

[root@########:~] cd var/run/log/
[root@#######:/var/run/log] cat vmkwarning.log | less
cat: read error: Input/output error
[root@#######:/var/run/log] cat vobd.log | less
cat: can't open 'vobd.log': Input/output error

2. Run the command "esxcfg-scsidevs -c" to check the devices connected to the ESXi. 

[root@########] esxcfg-scsidevs -c
Device UID                      Device Type                    Console Device                                                                         Size                            Multipath PluginDisplay Name
naa.################   Enclosure Svc Dev   /vmfs/devices/genscsi/naa.################                              0MB       HPP            Local Cisco Enclosure Svc Dev (naa.################)

3. Run hexdump against the device to test whether raw data on the device can be read. 

Example: 

[root@#######] hexdump -C /vmfs/devices/disks/naa..############:1 | less
hexdump: /vmfs/devices/disks/naa..############:1: No such device
hexdump: /vmfs/devices/disks/naa..############:1: Bad file descriptor

Resolution

Once RAID rebuild is confirmed complete, reboot the ESXi host.