You are not logged in.
I have a somewhat dumb question.
If I make an image of a HDD using dd/ddrescue then compress the image it should still be "entact" meaning if uncompressed it will return to a dd-able source image?
I was googling this ... and it was giving me results about jpeg image compression which isn't what I need. I realize I could try this by hand but dd/ddrescue are a little slow, and I only get so much time at the office.
The intended goal is to have an image of the physical drive, all the partitions (dd/ddrescue are good for this), and then make a compressed version of the file for archiving purposes(either a zip or a 7z).
Again I'm fairly sure this is "true" but I'm hoping someone else has some prior experience with something similar.
Offline
Yes, the image should decompress without corruption.
Is there a particular reason you need a disk image?
I prefer rsync(1) for backing up, the underlying partition table (and filesystem) becomes irrelevant if that tool is used and the system can be re-created in different configurations.
ddrescue(1) is for use on damaged drives so unless this is the case, dd(1) should probably be preferred.
Offline
Is there a particular reason you need a disk image?
I don't need a disk image but I've done, slightly, more with dd/ddrescue than rsync. That said I'm on board with doing rsync.
Another reason I'm angling more toward disk image is that it's a Windows HDD with recovery partitions, and thus there is no optical media from which to recover the OS install - a tactic I'm not particulary fond of. Thus if the image is "complete" I can put everything back where it originally expected to be so doing recovery from the drive itself to itself isn't ardueous.
hopefully that all makes sense.
Offline
For the benefit of the studio audience, an interesting reply from @dilberts_left_nut over at forums.debian.net:
zero-filling your free space prior to imaging will greatly increase the 'compressibility' of the resulting file
Offline
Also for the benefit of the studio audience.... Does anyone use the forked command of dd, dcfldd ?
http://linux.die.net/man/1/dcfldd
dcfldd is a fork from the Department of Defense and offers a bit more in terms of actual visual output (gives hash marks as a progress bar)
No one seems to mention dcfldd, plus i had no idea it was an acronym "Department of Defense's Computer Forensics Lab dd"
"I have not failed, I have found 10,000 ways that will not work" -Edison
Offline
^ I used to use dcfldd quite a lot - especially useful to see the progress. But I kept forgetting the command....er, is it "ddfldc", nope; let's try "dflcdd", nope... 8o
Be Excellent to Each Other...
The Bunsenlabs Lithium Desktop » Here
FORUM RULES and posting guidelines «» Help page for forum post formatting
Artwork on DeviantArt «» BunsenLabs on DeviantArt
Offline
^ I used to use dcfldd quite a lot - especially useful to see the progress. But I kept forgetting the command....er, is it "ddfldc", nope; let's try "dflcdd", nope... 8o
lol, yea it is a odd, seeming random bunch of characters.... but hopefully the acronym will help you remember it. I just think of it as DOD's Computer Forensic Lab's DD
That or...
Don't Copy, For Legal Decisions Dummy! < --- I made that up....
"I have not failed, I have found 10,000 ways that will not work" -Edison
Offline
That's really interesting! Thanks to you all for your input!
Offline
Don't Copy, For Legal Decisions Dummy! < --- I made that up....
Delinquent Cute Foreigners Love Domestic Diners
Debian 12 Beardog, SoxDog and still a Conky 1.9er
Offline