A VMFS volume that reports I/O errors can still be usable - if the I/O errors is located in unused areas of the volume.
If the errors occurs in the first 1200 gbs the volume will very likely be unreadable with ESXi itself.
I just noticed that I/O errors that lock a volume occur surprisingly often at 20 MB into the disk. Thats where the Heartbeat-section of 4 MB starts.
Coincidence ? - dont think so ...
Diagnosis:
dd if=/dev/disks/partition bs=1M count=2048 of=headerdump.2048
reports error and location/offset
Suggested Cure:
dont waste time with trying to fix I/O errors when they occur in metadata range - ESXi does not have the tools to continue - switch to Linux
dump partition and extract data from image
important: reduce errorsize during dump
Ulli
________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...