Client was hosting a Windows 10 Virtual Machine on XenServer, using a Jetstor SAN for the storage, the Jetstor uses fiber channel to connect to the host server, which is where all the VM’s reside; the Jetstor is a 16 bay unit, with 8 Mirrored 2TB arrays. Attached is a pic of the original RAID configuration.
Client logged into a physical Windows Server that has a fiber channel card in it, and could see all of the Jetstor’s on the network through its fiber channel NIC.
He plugged in a 10TB USB 3.0 external drive to the Windows Server, opened Disk Management to format it and a prompt came up to initialize the drive. The drive it was asking about was actually one of the 2TB mirrored arrays from the Jetstor SAN, not the 10TB USB 3.0 drive (In normal operation, all the 2TB RAIDS show up as un-initialized in Disk Management).
Thinking it was the 10TB drive he initialized it as GPT which then made the XenServer no longer able to access the file structure it had created on that array.
So far I've used RStudio, GDB, and Reclaime for logical scans. Also testdisk to try to find the lost partitions.
I've made multiple copies of each drive and spread the scans across a couple different machines to try to expedite the process.
On each individual drive, Reclaime is the only program that returns the entire folder structure but everything is corrupt and HEX data is garbage. The RAW information is somewhat intact but for example, one of the folders in the original structure is "employee pictures" with about 50 pictures that are all corrupt. I found the corresponding folder in the RAW recovery and some of them are OK but some are chopped in half vertically.
I have the SAN to work with. I have taken copies of the drives and re-initialized them with the original RAID parameters in the SAN and run the scans on that, no success.
The only thing I did that seemed hopeful for a minute is with the drives I initialized in the SAN with the original parameters, I created a Virtual Mirror with RStudio. There was one order I placed the drives in that showed an EFI partition and the original 1TB partition that was on the RAID before starting a scan. The result of the scan was exactly the same as a scan with an individual drive.
Testdisk detects the original 4 NTFS VM filesystems as HFS but unable to write them as NTFS.
It's pretty essential to reproduce an intact folder structure for this recovery and I'm out of ideas.
Sorry for the long post but wanted to include all details.
Client logged into a physical Windows Server that has a fiber channel card in it, and could see all of the Jetstor’s on the network through its fiber channel NIC.
He plugged in a 10TB USB 3.0 external drive to the Windows Server, opened Disk Management to format it and a prompt came up to initialize the drive. The drive it was asking about was actually one of the 2TB mirrored arrays from the Jetstor SAN, not the 10TB USB 3.0 drive (In normal operation, all the 2TB RAIDS show up as un-initialized in Disk Management).
Thinking it was the 10TB drive he initialized it as GPT which then made the XenServer no longer able to access the file structure it had created on that array.
So far I've used RStudio, GDB, and Reclaime for logical scans. Also testdisk to try to find the lost partitions.
I've made multiple copies of each drive and spread the scans across a couple different machines to try to expedite the process.
On each individual drive, Reclaime is the only program that returns the entire folder structure but everything is corrupt and HEX data is garbage. The RAW information is somewhat intact but for example, one of the folders in the original structure is "employee pictures" with about 50 pictures that are all corrupt. I found the corresponding folder in the RAW recovery and some of them are OK but some are chopped in half vertically.
I have the SAN to work with. I have taken copies of the drives and re-initialized them with the original RAID parameters in the SAN and run the scans on that, no success.
The only thing I did that seemed hopeful for a minute is with the drives I initialized in the SAN with the original parameters, I created a Virtual Mirror with RStudio. There was one order I placed the drives in that showed an EFI partition and the original 1TB partition that was on the RAID before starting a scan. The result of the scan was exactly the same as a scan with an individual drive.
Testdisk detects the original 4 NTFS VM filesystems as HFS but unable to write them as NTFS.
It's pretty essential to reproduce an intact folder structure for this recovery and I'm out of ideas.
Sorry for the long post but wanted to include all details.
