How to partition a GTP drive as separate devices please?

Rocky Linux Help & Support
1

Dear Rockers,

Would like to partition 1 ssd:

  1. I have a 4TB ssd that i need to split into 2 or 3 partitions.
  2. However i want the partitions recognised as separate devices.
  3. Rather than /dev/sdb1, /dev/sdb2 & /dev/sdb3
  4. I would prefer: /dev/sdb, /dev/sdc, & /dev/sdd
  5. Would like to copy data on different partitions but the backup software programs recognises the disk as one /dev/sdb.
  6. I would like to use 2 ntfs partitions to test … eg /dev/sdb & /dev/sdc (not /dev/sdb1 & /dev/sdb2).

HDD
HDD833×211 17.5 KB

Thank you

2

This seems to be the real issue: some “backup application” is limited to operate on block devices.
Shouldn’t the question then be how to make that application to behave (and not how to submit to its wishes)?

3

Thanks for the reply, all i need is to make the device path /dev/sdb separate, because the Rescuezilla overwrites all partitions on the drive under a single device path. Rescuezilla picks up other disk drives. I need to only copy data onto one partition.

4

Tools like Rescuezilla have their place, but have you looked at rsnapshot as a backup option? It uses hard links for unchaged files so a restore simply requires the date of the backup you want to restore from and a cp -rfp of that location to the original location. It’s not a flashy GUI interface, but really functional, fast, and resource conservative. If you are locked into Rescuezilla, no harm, no foul. I’m just throwing that out as an option.

5

Thank you for your reply, I need to do a clone & backup, which are different mechanisms thus the ssd drive is big enough for what i am trying to achieve. If i use Rescuezilla for cloning, i could then have the other partitions intact for other activities. I just need the other partitions ‘unscaved’ if possible. Is there a way on terminal (Linux) to create /dev/sda, /dev/sdb …on a same drive please?

Thanks

6

I think you don’t even understand your own question.

Want to clone ? Use dd wich works perfectly and can be used like you want.

Want juste to backup ? There are tons of programs which do it, best for me : tar

Take a moment to consider.

7

Thanks, “preposterous” :smiley: ; i know my question, but perhaps you don’t understand plus you didn’t offer a soln either…

  1. If you clone a disk drive both source & destination will be the same - right ?
  2. However if you want to clone & still have additional partitions appear as separate drives it won’t work or this feature doesn’t exist (unless you have a soln?).
  3. So again; this can’t be achieved → /dev/sdb1, /dev/sdb2 & /dev/sdb3 become /dev/sdb, /dev/sdc & /dev/sdd ? :smiley:

Forget the clone & backup, i can use the terminal however each partition can’t act as a separate device path on linux. That is the issue - hope it is clear.

I have found an rather unconventional manner of resolving this by cloning & then manually backing up (data to the cloned drive) to prevent the data being overwritten on unsed partition.

I will check with the slackware & gento community as they tend to compile everything; they might have some tricks.

Thanks

8

No wrong. Destination must be at least as big as the source. That’s all.

I will never try that. It is a non-sense. Or maybe you patch and compile your own kernel. Good luck with that.

9

Depends on what you talk about.

A clone is a verbatim copy of some data. So yes, it should have “same data”.

For example, the dd can read content of entire device, content of just one partition, or some part of them.
Where the dd writes to could also be an entire device, a partition, or a file.
Obviously, where you write to must have enough space to hold the data.

The dd could copy the content of entire device into a single file, into filesystem within a partition, if there is enough space for it. It could also copy from file into a device (like we write ISO image into USB drive).

A device does (usually) contain both partition table and partition(s). If you do write to a device, then you overwrite whatever data that device has (at least the beginning, if written data is smaller than the destination device). Therefore, there could be “other data” after the written content on the larger destination, but that is more or less inaccessible.