You can dump the last few sectors using dd and look for "EFI PART" signature:
esxcfg-info -s | grep -i vmfs Better yet, use the hidden voma tool (VMFS Offline Metadata Analyzer) in read-only mode:
Good luck, and may your sector scans be clean. Have your own VMFS partition table horror story or recovery trick? Share it in the comments. vmfs partition table recovery
The Ultimate Guide to VMFS Partition Table Recovery: When Your Datastore Goes Dark
We've all felt that cold sweat moment. You log into vCenter or ESXi, look at your storage devices, and see a datastore marked as or simply "Invalid partition table." Your VMs are inaccessible. Your heart rate spikes. You can dump the last few sectors using
Before you panic or reach for the latest backup (assuming you have one—and you should), take a breath. In many cases, the data on your VMFS datastore is still perfectly intact. The problem is often just the partition table —the map that tells ESXi where the VMFS volume starts and ends on the raw LUN or disk.
dd if=/vmfs/devices/disks/naa.6001234567890 bs=512 count=1 skip=END_SECTOR_NUMBER | hexdump -C | grep "EFI PART" But skip math is error-prone. Instead, use partedUtil : The Ultimate Guide to VMFS Partition Table Recovery:
partedUtil get /vmfs/devices/disks/naa.6001234567890 If you see Unknown or an empty table, the partition table is missing. Use vmkfstools to scan for VMFS superblocks without relying on the partition table: