Using virtualbox + system-rescue-cd (live linux) + usb3.0-to-sata HDD adapter for preparing the drive and the wdrecovery-stuff CAN WORK if you power of the drive while still attached to the live-linux WITHOUT giving windows the oppurtunity to f**k up the partitions
Long story:
Hi there everyone,
i read a couple of threads here on how to replace a drive in a MyCloud or how to unbrick a device, e.g.
So i was confident that my plan (buy an empty case and 8TB wd red, setup the partitions, copy the wdrecovery and fire it up with the current firmware) will work smoothly and could be archieved in a couple of hours. In the end it works but not as easy as i thought. I encountered an error which was not that obvious for me. So i decided to write down my experience, in case others are struggeling with a similar problem and to provide new food for their thoughts.
The methods to setup the harddrive in a Mycloud and get it running with the current firmware (e.g. described in the threads mentioned above), could be summarized like:
Boot up a Linux system
Prepare some partitions on the drive following a particular schema, if they’re not present anyways
Copy the wdrecovery-files (keywords: uImage and uRamdisk) onto a particular partition OR copy a raw-image of Mycloud drive back onto the drive (which includes the software, e.g. see here)
I tested both methods, but i prefered (and in the end i used) the method with the wdrecovery-files, cause it feels more like setting up the device from scratch. Using an image could also be another source of error.
While reading all the threads in the WD community, a couple of times a Google Drive with a lot of information is mentioned and many thanks to user Fox_exe to gather all the information.
Hardware i bought and used to setup the Mycloud device:
An empty MyCloud case (P/N: WDBCTL0060HWT-10 … original 6TB Gen2 using FW=2.xx)
8TB WD Red drive (WD80EFZX)
Used OS environment:
While my main machine is running Windows 10 x64, i tried to boot an older laptop with a Live-Linux via USB but this fails. I thought about why not using Linux in a VM. So i used Virtualbox (current sw: 5.2.6-120293-Win) without the guest extensions and a ISO of SystemRescueCd (current version: 5.2.0 x86).
Setup VM:
OS type=Linux; Version=Other Linux (64-bit)
RAM=1024MB; Virtual hard disk with fixed size of 1GB (not required so could be less)
USB: enable USB controller; USB 3.0 (xHCI) controller - required so i can use my usb-sata-adapter at the USB3.0 ports of my machine and give the control of the device to the VM (error message: failed to attach the USB device to the virtual machine )
Start VM:
Attach the ISO to the optical drive, boot-up the VM and select Option #6 (start graphical environment) … wait till GUI is loaded
Create partitions following the commands in the instruction
Hint: Use the cmd ‘print device’ in parted in terminal to check you selected the correct drive before messing something up
Created partitions in GParted (including swap on sdX1)
Copy wdrecovery-files to sdX3: I used a USB stick to transfer them into the VM (attach USB stick to VM, mkdir and mount device to dir)
Unmount drive from mounting-point and DO NOT DETACH the drive from to VM back to Windows, cause this f***ed up the partitions somehow (maybe when discovering the new drive/partitions) an lead into errorcode=600 later on. I just turned off the power from the usb-sata-adapter
Upload firmware and set up storage
Attach the curcuitboard back onto the drive, attach network and power cable and fire it up
Wait some minutes until the mycloud has booted into “Safe Mode” (the blinking LED switchs from blue to red)
the device will use try to get an IP via DHCP (or if not avaible if think it will use the IP-range of zeroconf) and is accessible via Web-Gui (LED is still blinking red)
Log onto the Web-Gui and upload current firmware for Gen2 (for me it was My_Cloud_GLCR_2.30.181.bin)
Reload the Web-Gui to get to the known WD dashboard and start the initial setup process ‘first steps’ (accept license, set password for admin and so on)
This error occurs if i detach the drive from Linux back to Windows after creating the partitions and shutdown the power of the usb-sata-adapter after ejecting it under windows. This error occurs NOT if i power down the usb-sata-adapter while still attachted to the Linux-VM. It is possible to activate SSH before setting up the RAID stuff, so maybe it would be possible to fix this error via SSH on the device, but i didn’t tried so far.
It seems that pre-formatting all partitions to ext4 (except swap) could avoid the error.
This sounds great and all but is there ANY way that this can be performed on a Mac Book Pro running Sierra? I can’t seem to find anywhere on how to do all this for the Mac. If someone knows of a link that shows how to perform this on a Mac, PLEASE, post it here.
Sorry to revive this topic, but I hope someone is still online on this problem.
I tried everything to replace the damaged disk if my 4TB MyCloud gen2
-I copied a 4TB image and a 3TB image, but all I got was a red blinking light
-I used FOX_exe’s directions to connect via Telnet. I did it over 10 times. The partitioning goes OK, I load the original firmware, I manage to enter the GUI of the device. But when I restore factory defaults or when I change RAID mode it reports the error 600, goes to Hybernate and the disk is a mess. That means that if I connect the disk to gParted it all the partition that I had created are gone.
-I connected the disk to a laptop running VirtualBox. Created the partition using gparted, copied uImage and uRamdisk to the /mnt/root/boot dir of sdX3. And I turned off the power of the USB-to-SATA where my 4TB disk is connected. But nothing again.
The only thing I can imagine that is different, is the original firmware. I download the most recent “My_Cloud_GLCR_2.31.174.bin”. As I can see you use this file: “My_Cloud_GLCR_2.30.181.bin” . Do you think I should use the earlier version too? I do not want to destroy my board using an older firmware in case that is not alowed by the system
What should I do? I am copletely stuck.
Finally I did it.
The solution is a mix of the two methods
I created the partition directly on the Disk (system rescue CD in Virtualbox) and (pre)formated them to ext4 except from the swap partition. There was an error while creating the sda3—> I chose to continue.
After that I connected the disk to the board and boot it with usb stick (FOX_EXE)
Using telnet, I see that the partition table is not recognized (strange because gparted did it in both cases). I use “mklabel gpt” and then I see that the partitions appear formated.
I follow the rest and when I come to “mount /mnt/sda3 /mnt/usb” there is an error again. (invalid argument). So I use “mkfs.ext4 /dev/sda3” which completes OK and I can mount it.
I finish, reboot, find it in browser (you may need a 40sec reset), load the old firmware “My_Cloud_GLCR_2.30.181.bin” and success. Format completes and all is OK.
When I try to autoupdate the firmware it reports error. I manually update it and all seems OK. Still the auto udate reports a failure.
Thank you all. it was a very frustrating experience. I learned some things about unix. I am very dissapointed though from WD. My device was working for only 2.5 years and it failed. I lost a big collection of pfotos and videos.
I think that we are all guided to online storage.
i also have a 8TB that i want to install and i did the partitions
now im stuck on the quote thing.
do i need to copy the files : uimage , uimage-wdrecovery , uramdisk , urramdisk-wdrecovery into sdb3?
from where do i need to do that? im not a linux guy so i have no idea… anyway this is how perform things ,
1.installed the virtual box ,
2.made a VMDK and boot with systemrescue
3.get into Gpart and make all the partitions .
now what ?
I had this same question. Seems documentation is always lacking. Following Fox_exe’s instructions, (which hasn’t worked for me as of yet) the commands listed creates a folder in partition #3 of the drive you’re modifying (sdX3) and then the cp command puts two files in there. (uImage-wdrecovery, and uRamdisk-wdrecovery) HOWEVER, that same copy command renames both files to simply “uImage” and “uRamdisk”. I have tried just about every permutation of putting the files in that partition including just placing the files from the USB drive Fox_exe has us create into there with no difference.
Has anyone managed to make this work with a 2nd generation MyCloud single bay device already running Firmware 5.26.119?
If so, how did you do it?
I am trying to clone from a working original 3TB disk to a new WD Red Plus 6TB drive, and so far failed abysmally.
Have used internal SATA ports, external USB caddys, Windows 10 + AOEMI, Ubuntu, and Clonezilla on USB environments, but all methods seem to have failed so far.
Both the old and new drives are physically fine - passing both short and long DST’s, and good on CrystalDiskInfo - so not worried about hardware problems, just can’t get it to work at a software level.
Sometimes able to load My_Cloud_GLCR_2.31.204.bin (original), not later versions, from emergency V1.xx firmware, but any attempt to get past JBOD formatting, and thereafter updating the firmware fails.
All help or suggestions much appreciated.
My apologies for reviving this topic but I’m stumped. I have a My Cloud 2nd Gen 2TB NAS. I have followed every step within FOX-EXE’s instructions and I’m able to upload the firmware and whatnot. The problem I seem to be facing is when I setup RAID JBOD. The system conducts a drive check and claims it’s good. After it being on FORMATTING VOLUME_1 for about 5 minutes, I’m met with error 635 and forcing me to hibernate the system. I have attempted to mess around with setting up sda2 with raid on using SET 2 RAID ON within parted but that didn’t work either.
