New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
boot to rescue
dd entire disk into over network to second machine
Boot to rescue and use ddrescue not plain DD as it will not stall on bad blocks so takes less time for better recovery chances.
if the drive can't be detected with lsblk does it mean it's not recoverable?
there is tool in hirens boot cd that can scan for partition(s) and recover their partitions mapping so lsblk can see it. but it might be in mbr scheme only, don't know if it had gpt support, so don't get your hope high yet. forgot their name, let me check on their list of dozens of apps.
use hirens cd 9.x which include all hacked licenseregistered apps first; if it's freeware, you can check the latest hirens cd.
found it, it's called "partition find and mount". it's freeware so you can use latest hirens bootcd.
there are also "Smart Partition Recovery" (freeware) and "DiskGenius" (shareware).
good luck.
If the drive isn't detected at all i fear there's little you can do.
If it is detected though: Don't try anything fancy. Use ddrescue as mentioned by @backtogeek to create an image at a remote location (i figure you don't have any other storage in the box where you could work on the image - use something like
ddrescue if=/dev/sdX of=/dev/stdout bs=1M | gzip -9 | ssh user@remote 'gzip -d | dd if=/dev/stdin of=windisk.img bs=1M') since with any access to the disk the chances of it failing completely rise. Any recovery attempts should use (ideally a copy of) the resulting image.Dealing with the image is probably best done using
losetup(man page), which will give you an actual device that's backed by your image. Depending on how damaged the Windows partition is you might be lucky enough to be able to just mount it using ntfs-3g (man page).is it an HDD or NVMe ?
It is SATA HDD
hmm, sometimes nvme's come back after leaving the server powered off a while to cool down, sometimes in read only, but HDD... if its not showing up at all, I am afraid it is probably not going to be possible.
Maybe give someone experienced and trusted access to the server in rescue mode to triple check but you should plan for the worst.
If you dont see it in lsblk (or lspci for NVMe) anymore your board is gone. OVH will ship you the HDD at massive cost and you can have the PCB replaced by a specialised service.
Its often just ca capacitor or similar but you need an exact donor drive and a cleanroom.
The HDD as in the spinning rust is in 90% of cases perfectly fine unless it was dropped while on. It is way harder to kill modern drives than ever, hell even the 1.8" ZIF drives in iPods survived hundreds of falls.
Is it? I thought it was usually head crashes and lubrication drying up, and other generally physical issues.
[Electrolytic] capacitors are often the first components on a circuit board to fail in general, but they last a lot longer than the typical failing HDD and it'll never cause SMART reports.
If you see SMART yes its toast, if it just disappered chances with a donor are good
If you don't see the drive, then the only option is getting it physically.
The newer the model, the less are the chances that you'll be able to do anything yourself afterwards. Quite likely you'll need to contact a specialized company (that isn't cheap usually).
If the drive is detected, create an image, preferably with gddrescue (GNU ddrescue).
Things are more complicated these days, unfortunately.
Capacitor (and other elements') issues are rare. You probably meant a 12V TVS diode, back in the days it was possible to remove it and replace with a bridge.
PCB issue is more realistic. Just a PCB swap is unlikely to help. If that's the case, then most likely at least ROM transfer is also needed.
Clean room required for a head swap, but that's way beyond even an advanced user's capabilities. Professional help is necessary.
Important P.S. (to OP and other interested)
Following an "advice" from youtube video with some Indian dude doing a head swap or even just opening a drive at home is a straight way to disaster.
If you're unsure about the issue, better to avoid even powering the HDD on.
Zero chance. You fail at the Helium refill even if you manage to not fuck up before.
I only have seen it in videos but the PCB swap seems to be rather common.
Of course for (NVMe) SSDs and eHDDs with integrated encryption you can absolutely forget any repair if your controller is toast.
Helium drives are recoverable (not each and every though - newest models aren't well-researched yet), but definitely more expensive than the "regular" ones. Some models are way more expensive.
Some models can read (not write) without helium, for others there are specialized boxes, which can be filled with helium (ain't cheap stuff either).
Just a swap was feasible years ago. Most relatively modern drives are unique (calibrated individually at the factory).
Google for adaptives (adaptive parameters). Without those HDD won't be able to read.
Hence the need for ROM transfer.
SATA SSDs and NVMe ones are two very different stories
Some "encryptions" can be decrypted, but that's at least a separate article as well.
Also vice versa, unfortunately.
The newer the drive, the more fragile it is. Miniaturization + more complex technologies under the hood.
SMR HDDs can just kill themselves during some relatively typical activities. They also may not survive incidents, which would be mostly OK for a CMR drive.
Oh g-d dont remind me, we had hundreds of the first gen SMR drives just randomly die. It was so bad they shipped you a new HDD even out of warranty, afraid of class action suit.