Step 6: Re-register each VM from the configuration file: # find /vmfs/volumes/Prod-HighSpeed -name "*.vmx" -exec vim-cmd solo/registervm {} \;
Within 90 minutes, 34 of 37 VMs booted cleanly. Three had corrupted swap files—recreated them. The fraud detection system was back online at 4:15 AM, just before the London trading desk opened.
The datastore reappeared in the vSphere Client. VMs showed as "unknown"—expected. She browsed the datastore: all VM folders, .vmdk , .vmx files intact. recover vmfs datastore
Step 5: Mount attempt on ESXi: # esxcli storage vmfs snapshot mount -l Prod-HighSpeed
Maya stared at the now-green dashboard. Somewhere in the datacenter, a disk blinked its steady heartbeat. She smiled. Another VMFS ghost story, beaten by knowing exactly where VMware hides its backup superblocks. Step 6: Re-register each VM from the configuration
Step 1: Identify the device. fdisk -l showed /dev/sde as 12 TB, with partition 1 (VMFS) starting at sector 2048. Good—partition still there.
Maya grabbed coffee and her battle-tested Linux VM with vmfs-tools compiled. First rule of VMFS recovery: do not write anything to the affected LUN . She used a rescue Linux live CD on a physical host with HBA access. The datastore reappeared in the vSphere Client
Post-mortem? They automated LUN binding policies, restricted SAN reconfiguration rights, and made the intern write a 10-page essay on SCSI device IDs.