HDD GURU FORUMS

Hello,
I have an SSD drive that using kind of Linux. It is an SSD drive for MITEL 3300 switch:
I want to create a backup from it. The SSD drive has below structure:

Code:

$ lsblk -f
NAME FSTYPE LABEL UUID                                 FSAVAIL FSUSE% MOUNTPOINT

sdc                                                                   
├─sdc1
│    vfat   /sysro
│                 020C-0000                                           
├─sdc2
│    vfat   /db   0257-0000                                           
├─sdc3
│    vfat   /vmail
│                 02F6-0000                                           
├─sdc4
│                                                                     
├─sdc5
│    vfat   /sysro
│                 036A-0000                                           
├─sdc6
│    vfat   /db   037F-0000                                           
└─sdc7
     vfat   /unused
                  03FA-0000           


First, I unmounted all of the partitions:

Code:

$ sudo umount /dev/sdcX


Then, used below command to create a clone:

Code:

$ sudo dd if=/dev/sdc of=/home/jason/SSD.dd
234441648+0 records in
234441648+0 records out
120034123776 bytes (120 GB, 112 GiB) copied, 2850.17 s, 42.1 MB/s


And:

Code:

$ file SSD.dd
SSD.dd: DOS/MBR boot sector; partition 1 : ID=0x6, active, start-CHS (0x0,1,1), end-CHS (0x8f,254,63), startsector 63, 2313297 sectors; partition 2 : ID=0xb, start-CHS (0x90,0,1), end-CHS (0x3ea,254,63), startsector 2313360, 13799835 sectors; partition 3 : ID=0xc, start-CHS (0x3eb,0,1), end-CHS (0x3ff,254,63), startsector 16113195, 29543535 sectors; partition 4 : ID=0xf, start-CHS (0x3ff,0,1), end-CHS (0x3ff,254,63), startsector 45656730, 188784918 sectors


After it, I plugged another SSD drive with the same capacity and executed below command to recover the image to this new SSD drive:

Code:

# dd if=SSD.dd | pv | dd of=/dev/sdc bs=1M
234441648+0 records iniB/s] [                      <=>                         ]
234441648+0 records out
120034123776 bytes (120 GB, 112 GiB) copied, 4376.66 s, 27.4 MB/s
111GiB 1:12:56 [26.2MiB/s] [                     <=>                          ]
1+2129501 records in
1+2129501 records out
120034123776 bytes (120 GB, 112 GiB) copied, 4415.03 s, 27.2 MB/s


But, when I plugged this new SSD to the switch then it doesn't work!
Any special tools for cloning this kind of operating system?

Thank you.

The operating system is irrelevant to the cloning procedure.

Just do a sector-by-sector clone to another hdd using HDDRawCopyTool (win) or HDDSuperClone (linux).
As you're already using linux, HDDSuperClone will be your pick.

Connect destination drive, launch HDDSC, select source and destination and just let it finish.

northwind wrote:The operating system is irrelevant to the cloning procedure.

Just do a sector-by-sector clone to another hdd using HDDRawCopyTool (win) or HDDSuperClone (linux).
As you're already using linux, HDDSuperClone will be your pick.

Connect destination drive, launch HDDSC, select source and destination and just let it finish.

I already did it!, "dd" don't do it?
What is the different between "dd" and "HDDSC"?

HDDSuperClone Pro is a very powerful cloning tool. Try the free version (somewhat limited) or by a 2 month Pro licence for not very much. As you already understand Linux you'll soon be up to speed.
http://www.hddsuperclone.com/sitev1/

hack3rcon wrote:

northwind wrote:The operating system is irrelevant to the cloning procedure.

Just do a sector-by-sector clone to another hdd using HDDRawCopyTool (win) or HDDSuperClone (linux).
As you're already using linux, HDDSuperClone will be your pick.

Connect destination drive, launch HDDSC, select source and destination and just let it finish.

I already did it!, "dd" don't do it?
What is the different between "dd" and "HDDSC"?

I actually did a test with the dd commands, and the outcome was an exact duplicate of the device. So unless something went wrong, there may be something else going on. Maybe the switch it somehow tied to the serial number of the drive. Or maybe the drive has an HPA or DCO where the switch is accessing a hidden area that is not cloned by normal methods without removing the HPA or DCO.

One thing I will point out is that using pv with the dd command doubled the amount of time it should have taken to write back to the drive.

There doesn't appear to be a HPA. Using the IDEMA standard, the reported number of LBAs is consistent with a 120GB drive.

97696368 + (1953504 x (120 – 50)) = 234441648

FYI the MITEL 3300 is not a switch, it is a proprietary VOIP controller. It would not surprise me if it is locked to the drive somehow. And if you mess up the original drive, you may brick the whole thing.

A quick Google search told me that Mitel requires the replacement drive to be of the size and manufacturer at least. So it could be that.

maximus wrote:

hack3rcon wrote:

northwind wrote:The operating system is irrelevant to the cloning procedure.

Just do a sector-by-sector clone to another hdd using HDDRawCopyTool (win) or HDDSuperClone (linux).
As you're already using linux, HDDSuperClone will be your pick.

Connect destination drive, launch HDDSC, select source and destination and just let it finish.

I already did it!, "dd" don't do it?
What is the different between "dd" and "HDDSC"?

I actually did a test with the dd commands, and the outcome was an exact duplicate of the device. So unless something went wrong, there may be something else going on. Maybe the switch it somehow tied to the serial number of the drive. Or maybe the drive has an HPA or DCO where the switch is accessing a hidden area that is not cloned by normal methods without removing the HPA or DCO.

One thing I will point out is that using pv with the dd command doubled the amount of time it should have taken to write back to the drive.

Can I copy the serial number of the old SSD to the new one?
How can I sure about HPA or DCO?
Others cloned the SSD drive by HDD Docking device like https://www.jaycar.com.au/medias/sys_master/images/images/9369001590814/XC4697-dual-2-5-3-5-sata-hdd-docking-station-usb-3-0ImageMain-515.jpg and its worked!

The HDDSC is not OK for bootable Disks?

hack3rcon wrote:The HDDSC is not OK for bootable Disks?

HDDSuperClone is very good for bootable disks. It makes an exact sector by sector clone/image of the whole disk.

Mitel has a controller that checks for drive serial number and brand. That it's locked.

DRUG wrote:Mitel has a controller that checks for drive serial number and brand. That it's locked.

Then, How a HDD Docking device did a clone and it's worked?

How can I compare the contents of two HDD sector by sector?
I checked the MD5 of Original SSD Drive with Cloned SSD Drive and all partitions has different MD5 codes.

What is the different between "dd" and "HDDSC"?

HDDSuperClone is specifically designed to handle defective drives, which is not the case here, so to clone or image a non defective drive they should have the exact same outcome, I don't know why HDDSC was suggested here.

How can I compare the contents of two HDD sector by sector?
I checked the MD5 of Original SSD Drive with Cloned SSD Drive and all partitions has different MD5 codes.

As it's been replied already in that other thread, it can be done with WinHex, which you apparently already did, and found some different bytes. Now the question is : what are those different bytes ? Could be that the partitions' IDs were updated when both the original drive and the clone were plugged at the same time, to avoid a conflict (which wouldn't happen in a docking station performing a clone autonomously, without being plugged to a computer), and the device in question stubbornly wants the original IDs. (Not sure about that at all, no experience with anything that looks like this thing — to say it in Bill Gates idiot-proof english, “the thing that solves the thing”.)

One thing I will point out is that using pv with the dd command doubled the amount of time it should have taken to write back to the drive.

What does that “pv” switch do ?

Or maybe the drive has an HPA or DCO

What do HPA and DCO stand for ?

What does that “pv” switch do ?

It is not a switch, but something else. Do what I had to do to find out what it was. Do a (Google) search for "linux dd pv" and look through the results.

What do HPA and DCO stand for ?

Again, (Google) search is your friend, "hard drive hpa dco".

I put (Google) in parentheses because that is the search engine I used. You should also be able to use other search engines to find the results (#ihatebing).

Quote:
What is the different between "dd" and "HDDSC"?

HDDSuperClone is specifically designed to handle defective drives, which is not the case here, so to clone or image a non defective drive they should have the exact same outcome, I don't know why HDDSC was suggested here.

Honestly, when I want to do something with dd, I use ddrescue instead, partly because I am already familiar with the command format, but also because it already shows the progress, so no need for pv.

Again, (Google) search is your friend

Well, first, no, Google is not mah friend, it's nobody's friend except those who run Google and those who benefit from their partnership with Google, and moreover, as you might know, doing a search on any Web search engine can lead to the right answer right away, but can also more often than not lead to a tremendous amount of B.S., or in between, answers which are technically correct but involve a prerequisite knowledge of the subject to be clearly understood, or technically correct but formulated in such convoluted or confusing terms that without an already comprehensive knowledge of the subject it's going to be misleading to anyone aiming at actually learning something, and that's why forums like this one exist, so that people of various knowledge and experience level can interact directly and collectively to quickly get to the point and separate facts from B.S. in a timely manner. {1} Perhaps the answer to those two questions is simple, from the point of view of someone who already knows it, but, since I've been reading quite a bit on the topic of HDDs and data recovery over the past few years, when I read something that I had never seen mentioned anywhere before, I tend to figure that it's quite advanced or arcane stuff, and, given the opportunity, prefer to ask someone who can provide a straightforward reply — in about as many characters as it requires to advise to go google it, possibly less — rather than going into yet another potential fool's errand, which I do too much already {2}, and which is highly likely given how the global signal-to-noise ratio of our world is dwindling on a yearly basis.
(Take that Marcel Proust and James Joyce. I'll reply in haiku next time.)

Honestly, when I want to do something with dd, I use ddrescue instead, partly because I am already familiar with the command format, but also because it already shows the progress, so no need for pv.

I, too, prefer to use ddrescue rather than dd with non defective devices (because more familiar with the syntax as well, and it provides useful options like “sparse” writing), but the point here was that the O.P. used dd, which should have worked well for the intended purpose, but did not, and therefore it's unlikely that using ddrescue or HDDSuperClone instead can fix the issue, since, for the intended purpose, they do nothing more.


{1} Take a breath, brace for second sentence.
{2} One recent example : yesterday, while translating subtitles for this movie (which I highly recommend by the way), I searched what “SW Keihin” stood for, with those keywords all I got was a manufacture of cars and motorbikes carburetors founded in 1956, which could not be relevant to 1945 Nagasaki ; then I tried “SW acronym”, and with a 2 letters acronym you can bet that I got a lot of irrelevant B.S., the first hit on page 1 being “sex worker” from the never disappointing Urban Dictionary... After about half an hour, I got some hints that “SW” might stand for “Shinjuku Ward”, but then I couldn't find an obvious association with “Keihin”, so I gave up, and just put “Shinjuku”, which may or may not be true — and noone will care whatsoever when watching the movie anyway.

abolibibelot wrote:

Again, (Google) search is your friend

Well, first, no, Google is not mah friend, it's nobody's friend except those who run Google and those who benefit from their partnership with Google, and moreover, as you might know, doing a search on any Web search engine can lead to the right answer right away, but can also more often than not lead to a tremendous amount of B.S., or in between, answers which are technically correct but involve a prerequisite knowledge of the subject to be clearly understood, or technically correct but formulated in such convoluted or confusing terms that without an already comprehensive knowledge of the subject it's going to be misleading to anyone aiming at actually learning something, and that's why forums like this one exist, so that people of various knowledge and experience level can interact directly and collectively to quickly get to the point and separate facts from B.S. in a timely manner. {1} Perhaps the answer to those two questions is simple, from the point of view of someone who already knows it, but, since I've been reading quite a bit on the topic of HDDs and data recovery over the past few years, when I read something that I had never seen mentioned anywhere before, I tend to figure that it's quite advanced or arcane stuff, and, given the opportunity, prefer to ask someone who can provide a straightforward reply — in about as many characters as it requires to advise to go google it, possibly less — rather than going into yet another potential fool's errand, which I do too much already {2}, and which is highly likely given how the global signal-to-noise ratio of our world is dwindling on a yearly basis.
(Take that Marcel Proust and James Joyce. I'll reply in haiku next time.)

Honestly, when I want to do something with dd, I use ddrescue instead, partly because I am already familiar with the command format, but also because it already shows the progress, so no need for pv.

I, too, prefer to use ddrescue rather than dd with non defective devices (because more familiar with the syntax as well, and it provides useful options like “sparse” writing), but the point here was that the O.P. used dd, which should have worked well for the intended purpose, but did not, and therefore it's unlikely that using ddrescue or HDDSuperClone instead can fix the issue, since, for the intended purpose, they do nothing more.


{1} Take a breath, brace for second sentence.
{2} One recent example : yesterday, while translating subtitles for this movie (which I highly recommend by the way), I searched what “SW Keihin” stood for, with those keywords all I got was a manufacture of cars and motorbikes carburetors founded in 1956, which could not be relevant to 1945 Nagasaki ; then I tried “SW acronym”, and with a 2 letters acronym you can bet that I got a lot of irrelevant B.S., the first hit on page 1 being “sex worker” from the never disappointing Urban Dictionary... After about half an hour, I got some hints that “SW” might stand for “Shinjuku Ward”, but then I couldn't find an obvious association with “Keihin”, so I gave up, and just put “Shinjuku”, which may or may not be true — and noone will care whatsoever when watching the movie anyway.

I answered with search terms because that is how I look things up to understand. I had to look up pv, and after my original reply I honestly didn't remember exactly what it was until I looked it up again after reading this post. And is is not something I can explain any better than the search results. My answering with search criteria means that I am not willing to take time to try to interpret all of the possibly search results you may get into something that you may or may not understand, when I myself may or may not fully understand it, and therefor have no chance of explaining it to anyone else.

Seriously, if you can't search or are afraid to search for how things work, then you will be very limited in what you can accomplish. You have no idea the level of things I have to search to be able to accomplish what I do.

and that's why forums like this one exist

Let me humbly disagree...

people with high level of knowledge are not here to answer basic questions. Everybody has a lot of stuff to do and it is not answering the trivial.
We did our homework too, and we also do it before asking on a forum, which takes time to read and write answers.
Of course i understand your point, that this is slightly less effective than asking a human being, but still, this should be the first step coz you may find the answer yourself, maybe with somewhat more efforts, but without taking time from others.
We must respect the time of other members and not use them as a search engine.

pepe