How to use Ddrescue to get data from a failing drive

This is a basic guide for using Ddrescue to get data from the most data from a failing drive and making a image of it.

[@mosbuk8107]

There are LOADS of ddrescue tutorials on here, but this is BY FAR the BEST one. I know, I watched them all. Pretty much the only one that goes through the process END-to-END. All others cop-out once they've issued the ddrescue command and fail to show their results which is pretty useless to be honest. Thank you EW. Much appreciated.

[@HellScream107]

You are truly the Electric Wizard. This worked for me today. I was able to recover a data from a dying harddrive. The current version of Ubuntu (24-04) actually has a built in tool to mount .img files. Thanks for the help!

[@onjisan]

BEST TUTORIAL EVER. Man you saved me! Thought I would never be able to get the data off of this SSD that couldn't be recognized by anything! Thank you!

[@TheRainbowPilot]

Excellent thanks for the help! Did not think I would be able to recover this drive, now i've got a stack of "dead" drives to go through haha. Thanks for posting this guide!

[@sohappywhenxyousmile]

You are a lifesaver!! Windows couldn't do anything with my SSD but ddrescue worked perfectly. Great tutorial.

[@iCarlyfan4534]

you are literally the greatest of all time, this might be the most specifically helpful tutorial ive ever followed. thank you so much

[@MagistusVerre]

Worked great with only some small issues, no other method worked. Thank you loads. Saved me 1TB of data after many days of hassle.

[@jtapoling]

Great info. I watched about five other videos before finally successfully running ddrescue. There is a lot of either bad or dated informational videos on KZbin. The drag and drop folder location was a huge help. This worked for me on Ubuntu 18.04.1 LTS. Thank you for posting this.

[@NetRolller3D]

8:34 It's failing to mount because Windows' built-in image mounter only supports partition images, beginning with a boot sector (or in Linux parlance, a superblock), while this is a drive image with a partition table at the start. ImDisk can mount this type of image just fine, but not the Windows built-in tool.

[@bronney]

Dude, you just saved me at least $300. Well done sir!!! Just a quick note, if the mobo doesn't boot the usb, click advanced in rufus and check fix for old bios when creating the image, did it for me. Thank you so much.

[@jkerman5113]

Thank you so much for this detailed tutorial. Compared to the other shit I've tried for this, DDRescue is art. Note: I initially tried to put the ubuntu file system as NTFS in Rufus, but the USB wasn't recognized. Changed it to fat32 and things went fine.

[@jlwhite]

Your guide still works even after 5 years thank you

[@hrishikeshparab1236]

Thats good to know cause my harddisk has also failed...

[@jlwhite]

@@hrishikeshparab1236 I wish you the best of luck recovering your data; I caught my drive going bad somewhat early, so I think I will be able to save the data so far; I'm at 99.88%.

[@hrishikeshparab1236]

@@jlwhite Thats good to hear , I have found myself in a similar situation too. I used foremost to recover some pdfs and jpgs. But I found out it’s better to create an image of the dying HD using ddrescue and then foremost it .I also tried photorec but that didn’t help.All the best to you too!

[@patpattson]

Thanks alot! I finally know what to do with the .img-file. Haven't found a single tutorial on how to get the files out of the image file but thanks to you sir now I know! :D

[@Psychiatry.321]

i use ddrescue on macOS on Mac Pro 2010 (Desktop) and it is awesome i always connect the bad drives to the sata ports and use /dev/rdiskX (by using r) prior the disk integer rather than /dev/diskX it will be like 15 times faster on Mac. also you can use Mac OS X El Capitan for better compatibility and it will be better than Linux (fundamentally the same). Thank You For This Video

[@NetRolller3D]

7:24 It's recommended to at least let the trimming and scraping phases finish before stopping ddrescue.

[@Jorge-hz6hp]

Great guide, Thanks.

[@jessesanchez5294]

Thank you!

[@Josivis]

OS runs fast then I hear a ticking sound from HDD then the OS slows down in Process viewer I see a 100% active time but no proccesses us using disk.

[@CoconutPete]

great channel! I'm learning the hard way how to recover data.. sadly through trial and error. There's a lot of bad information out there, that's for sure lol

[@chiliphil64]

Ddrescue also works on OSX

[@TetrisToasty]

How do you resume the scanning process (I.e. you have a logfile) when you've completely exited the terminal to test the recovery? I wanted to see if I could access the files recovered on the .img thus far, before the process was complete (it was at something like 99.21%), but no bueno. Presumably I have to boot Linux up again and actually let ddrescue fully complete the scan. How do I resume it? I'm worried if I type exactly same stuff in again from the video, the 99% complete log file will get overwritten and the scan will start from the beginning. Great video, by the way! Very helpful!

[@TetrisToasty]

Okay, I took a risk. To pickup the scan/recovery from where you left off, you DO punch in the exact same stuff as in the video (even installing gddrescue if booting from a USB). Instead of typing in the directories to create the img and log files though, you drag the img file and the log file you already produced from the first attempt (which you may of paused or bailed from). If you did not create a log file the first time around...you will be starting your scan from the beginning again. The .log file is a file that is a record of where the scan was up to (which is useful if you have to pause or exit the process for a bit). I highly recommend following every step this guy (whos an absolute champion) outlines in this video.

[@viisteist1363]

how is all of that extraction going to work with a 1TB drive(you will need a 3TB backup?) and would it be better to just clone the whole thing in the first place?

[@FSX3000]

5:24 but on the "Device" description it says "/dev/sda1" not "/dev/sda" should we ignore the number ?

[@ElectronicsWizardry]

/dev/sda is the hard drive. /dev/sda1 is the first partition. This drive only had one partition so you can clone just sda1 here, but normally id clone the whole drive incase you want data from other partitions.

[@Josivis]

5:59 So I can use the log file as a bookmark to continue later, I don't have the luxury of doing recovery during the evening, Here in South Africa we are on level 6 of rolling blackout/load shedding meaning up to 8hours without electricity spread throughout the day sometimes 4hours per session.

[@ElectronicsWizardry]

Yea the log file should allow ddrescue to restart after a power outage. So I’d just start ddrescue with a log file and if power is lost start it up again when the power comes back again run the same command with the same block device and it should continue. I’d suggest running it on battery or ups if possible as sometimes a drive just won’t start back up again after a power loss.

[@Josivis]

@@ElectronicsWizardry I plan to stop the scan with ctrl-c 30min before the scheduled outage then shutting down the pc safely.

[@LandLos.]

this kinda worked, so I'm in the last step in windows 10, I extracted the files. but the hddimage file had disappeared for some reason. I've got a copy tho. not sure what to do now

[@LandLos.]

Ok so I found a fix to this I guess, basically just check whatever drive you put this hddimage file in folder size and it should appear for some reason

[@surajdave545]

I cant extract cause the image is the same size of the disk :(

[@daneayers7483]

Thank you very much for this tutorial. Clear, Precise and easy to follow. I have a question for you or anyone reading this. My rescue is still going. Basically I have a 4TB Seagate nas drive that failed. I have another 4TB drive where I'm rescuing the data to. Once it's done will I have to uncompress the image twice like you did to get the data off? I don't have another two 4TB hard drives at hand. I've got some 2TB Nvme drives but that's only half the image size 🤔

[@ElectronicsWizardry]

Not having extra disk space makes this a bit more difficult. I'd try to DDrescue from the bad drive to the good drive. Just put the /dev/sdX as the output, not a file. Depending on the issue, moving to a drive without issues can fix the issues you have. I'd still recommend copying the files over in case the filesystem has issues due to the failing drive.

[@daneayers7483]

@@ElectronicsWizardry Thanks for the reply mate. So here's where my issues start. I had to clear out a 4TB hard drive so I had somewhere to copy the bad drive to. A seagate NAS is encrypted so nothing recognises it except ddrescue and HDD Raw Copy Tool for windows but thats useless because it takes forever and does not handle the bad blocks well. I'm using ubuntu-22.04.3-desktop-amd64 and while I was waiting I was watching a movie on youtube and when that finished I opened chrome in firefox and the system crashed near the end of copying. googled it and its a common problem with the new Ubuntu. So I fired it back up and did the exact same process now it will not start the transfer of files. I push enter and I just get a blinking > No matter what I did it just wont work. Not sure what the issue is. I need to start it again to get all the files off. The next issue is as its encrypted the drive structure is about 6 different ISOs one being 4TB where all my movies, series and music is. They are inside the ISO file. I could be wrong but Im thinking ill need another 4TB hard drive to extract the ISO to? Plus I cant get it to work again 😞 Any help/advice would be appreciated

[@Ajsski99]

Please help :( when I extract with 7Zip it says “Can not open the file as [NTFS] archive”

[@eurlmokrani2033]

Which system are you using ?

[@nazarenoalessandrolumia6587]

Hi, I have a problem at the beginning of the procedure. as soon as I run the command, it gives me the message "Can't open output file: not a directory" or "Can't open output file: no such file or directory". Please can you help me?

[@ElectronicsWizardry]

Can you show the command your using? I’d guess there is an issue where one of the paths you’re using is incorrect.

[@rGGdom]

will this work with a hard drive that now requires to be initialized?

[@thecurrychallenge]

My drive was failing and would be detected off and on. Going through your steps, Ubuntu Disks isn’t recognizing the drive that’s going bad. Is there another way to go continue or does that mean it’s toast?

[@theeducationalguy]

8:48 "Can not open as archive". Still did not found a solution other that photorec. App that does find my custom signatures.

[@Spectre.007]

can I use ddrescue in external hdd which was in raw state? and is unreadable and windows is asking to format the drive

[@indicator27]

i cloned from drive to drive what can i do with the raw file system on the new drive, the drive was dropped, testdisk fixed some stuff before ddrescue 99.9% cloned

[@DeathByBFG]

I think I may have missed something but what is the point of Rufus? I don't think it was used in the video.

[@ElectronicsWizardry]

I used it to make the Bootable Ubuntu usb I think. There are other methods but ddrescue needs to run on Linux and a Bootable USB is a simple way for most people.

[@DeathByBFG]

ElectronicsWizardry Thanks for your reply! I ended up running Ubuntu from a LiveCD and have DDRescue running right now.

[@ElectronicsWizardry]

Cool. Hope it works for you.

[@jalaj3524]

I'm getting error mounting my internal harddrive, it gives error - error mounting dev/sd1 in media mint system .... (Udisks-error-quark, 0) Please help

[@ElectronicsWizardry]

My guess is this would be the os is trying to auto mount the drive. Does ddrescue still work? Or is this after you made the image.

[@kbidols]

Hi there. I am currently using Ddrescue to recover data from a 500GB HDD. So far it's been running for 6 hours and it seems to be doing just fine. It has recovered 0.06% of data. However, it says that the remaining time is around 370 day - 410 day. Do you suggest me to keep on going or just stop right away since it's going to take literally forever to finish?

[@ElectronicsWizardry]

I'd give it Anouther day or two to try to recover the data. There is often one very slow part of the disk and the rest is much faster.

[@kbidols]

@@ElectronicsWizardry Okay then. I'll update you in a few days. Thanks.

[@kbidols]

@@ElectronicsWizardry Update: After 2d 22h 24m, the process suddenly stopped. The reason was because my PS2 keyboard suddenly shuts off. Here's the stats at that point: ipos: 3120 MB, non-trimmed: 196608 B, current rate: 0 B/s opos: 3120 MB, non-scraped: 0 B, average rate: 12253 B/s non-tried: 497001 MB, bad-sector: 0 B, error rate: 131 kB/s rescued: 3105 MB, bad areas: 0, run time: 2d 22h 24m pct rescued: 0.62%, read errors: 3, remaining time: 424d 2h What would you suggest me to do next?

[@esecallum]

@@kbidols try polishing the contacts of the hd board which connect with hd contacts. i have saved 18 drives by doing this. they are rusted or have oxide growing on them. i scrape them with a screwdriver and a slight polish and screw the board back

[@LewieLeway]

My laptop stopped booting. I've removed the HDD and loaded into a disc reader on my desk top. The HDD spins and both partitions show as disc G and I. Trying to access G I'm told the parameter is incorrect. Trying to access I and it says this folder is empty. Using Testdisk, G is showing as 968gb and when I scan it it just says read error at every sector. I've tried Recuva but that can't see anything. Is there any chance of this software recovering anything? Cheers.

[@ElectronicsWizardry]

FIrst thing make a image with ddrescue, you don't want to try to recover data from the disk, you want to make a image then recover data from the image. Then Depending on the data you may be able to just get the data from a disk image, or normally you will have to rebuild the partition table or using something like photorec that will recover data from known file headers and footers.

[@LewieLeway]

Thanks for the reply EW. I have been a total tit. In my frustration, before you had chance to reply, I decided to commit a sin and open the drive!! I even knew it was the wrong thing to do. It was a bit like tourettes. I couldn't stop myself. Anyway, all looked ok, I shut it again and then got the click of death, the head was flitting back and forth over the platters. Then the head stopped and hasn't moved since and it isn't recognised. The platters are still spinning. I take it I have murdered the beast beyond salvation? :-( As a matter of interest why does the head stop moving, circuitry?

[@duskfall_777]

It doesn’t see the drive

[@himalbheg]

Hi, i have recovered 37GB img data..but when i extracted it using 7zip.it doesn't show much data more than 20MB ...what should i do ?

[@ElectronicsWizardry]

How big was the drive you are trying to get data from. You might want to try other programs like photorec or Rstudio as there is too much data missing to get the filesystem working again

[@hippa2dahoppa2]

i downloaded ddrescue and it was like a boatload of files and no explanation how to use them. its like how they want you to guess. and theres no exe lol. its ridiculous.

[@ElectronicsWizardry]

I’m pretty sure ddrescue is a Linux only program so there is no exe for it. Did you boot into a Linux live disk to run ddrescue?

[@leocycy]

Nice tutorial thanks i need to ask if there is another way instead of .img to skip the 7 zip extract for example directly to a folder

[@GREGELZZ]

use imgburn to mount the image

[@skullrazor]

What if your drive hadn't failed, but have mistakenly installed Ubuntu over the drive, can the old data be recovered? It took about 5 mins to install Ubuntu and I have a 1tb drive of data which I think is still recoverable.. but anyone know How?

[@ElectronicsWizardry]

This method would not work in that case but id still make the image and then use a program like photorec. Installing Ubuntu got rid of the file system info on where the files are do photorec will search the drive for the begging and end of files. A data recovery service will do a bit better with this task if there is very important data on the drive.

[@skullrazor]

ElectronicsWizardry why wouldn't this method work? And if I did, how would I mount the image to then use photorec to recover? I used photorec to recover straight from drive but at bad sectors, the reading repeated in loop, and I heard using an image with ddrescue will resolve this issue or skip the bad sectors.. right?

[@bartekb5780]

good video. thanks. but a week of copying? I got a 2TB drive which just failed. How long will that take to copy? Daaamn

[@ElectronicsWizardry]

+Bartek B if the drive is fully dead then this method won't work. You will need to take the drive to a data recovery service. There are some tricks to get fully dead drives working again but I wouldn't suggest them if the data is valuable as they normally have a low success rate and can lower chances of a data recovery service if you decide to do that later. The copy speed is due to how the drive slows with bad sectors. The copying of the 2tb of good files will be fast. If in a time crunch you can stop after the first run and just not get the normally 100mb of bad data. That 100mb can normally get down to a few hundred kb if your willing to wait.

[@bartekb5780]

how can i determine is that drive "fully dead"? It would be enough for me to copy specific files from that drive who went broke but unlucky for me in this situation I had it whole encypted with TruCrypt so I assume I won't be able to view directories and files on that drive. Only whole disc clone/copy could work I assume. Also can you explain me briefly if I use ddrescue to clone that 2TB with 3 partisions drive to a newly bought 3TB drive for a backup will that 3TB drive include all the restored/copied data and a 1TB of unallocated drive space?

[@ElectronicsWizardry]

+Bartek B does the drive show up in the bios or disk management? If not it'a fully dead and this method won't work. For the drives is partition the 3tb drive like any other drive with ntfs and then make the image a file. You will then have 1tb of space on the drive to copy any other files to, like any other drive.

[@bartekb5780]

so before whole procedure I need to format this 3TB with NTFS am i getting it right?. Or just leave it as it was out of the box which is 3TB of unallocated space? And as to that broken drive. Windows disc management tool doesn't load up I don't know why. But I tried to clone that drive under Windows today with softwares like MiniTool Partision Wizard Free and HDD Raw Copy Tool and they both show up that broken drive. Also drives "monitoring" app I've installed called Crystal Disc Info shows up that disc but it just says "bad". That mentioned above MiniTool Partision Wizard Free app showed up that drive with 3 partisions on it before I made a try to clone it with that program but it failed and after this attempt it shows up that drive but without those 3 partisions like it did before. Now it just says "bad disk". Reasumming this drive is visible just without partisions on it so I assume it's not "fully dead" if it's still seen by drive management programs.

[@ElectronicsWizardry]

+Bartek B you want to format the 3tb drive as ntfs. Since it shows up in software it's not fully dead and you should still be able to make a image of it.

[@hemmatv1582]

Please learn how to articulate when you speak. Some words I dont understand/hear.