Possible to resume? Is img size of data or size of drive?

[conrad10781]

I'm currently running reallymine for the first time. Great product!

First question, if something fails, and the program has to be re-run, does it start from the beginning, or is it able to resume?

Second, I was curious if the img created from the decryption was the size of the data on the drive or the size of the drive itself.

I'm ~6 hours in and have 5.7GB in the img file. If it's based on data, that's not too bad of an issue as I believe there was ~75GB of data on there. If it's based on the entire drive size, it looks like I might be looking at over a month for this to complete.

I saw #38 , but didn't want to try this on my first run.

[andlabs]

I don't have a resume option set up yet. If #38 is confirmed, I could then modify it to provide a resume facility in the same fashion as GNU ddrescue... Try it anyway (if you need a binary I can provide it later), at least for the first 7GB that you have now; I would really like to know if I did make it faster, and if it still operates correctly (compare the 7GB you have now with the 7GB you produced; you can use --disk-size to make it stop early).

reallymine decrypts the disk as a whole, yes; this is the only decryption option that reallymine can do, because that's how the hard disk is encrypted. You can use the --disk-size option to make it stop early, but if you aren't sure about the partition layout of your disk (or if there is only one partition) this wouldn't work.

Hopefully one day we can transparently decrypt a mounted drive; I think themaddoctor (who lurks in this issue tracker) was working on that.

[conrad10781]

@andlabs , I would need a binary to test #38 , I've had issues building on an Ubuntu variant due to the IO error I saw in another ticket and that my release has Go 1.X , which I need to keep where it is for another piece of software at the moment.

According to iotop, the process is writing ~275K/s to the target drive , so it should be pretty easy to compare.

Does that branch have a way to limit the CPU's used? I have an older AMD FX(tm)-8350 Eight-Core Processor , and would like to use 4-6 cores to test if possible, in the event it still takes a while I need the machine to still be usable.

[andlabs]

The code I provided there uses all CPUs, but providing an option to restrict the number of CPUs is trivial. I'll hardcode it to 4 for the binary I give you later. What OS do you use normally, and what arch?

[conrad10781]

(L)Ubuntu, x86_64

…

On Thu, Jun 1, 2017 at 10:55 AM, Pietro Gagliardi ***@***.***> wrote: The code I provided there uses all CPUs, but providing an option to restrict the number of CPUs is trivial. I'll hardcode it to 4 for the binary I give you later. What OS do you use normally, and what arch? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#40 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAOkcRydTeBsgrOZEul1tQa6MjOdNcEYks5r_tD-gaJpZM4NscKy> .

-- Matthew C. Rice RCS http://rcs.us [email protected] TF - 844-4-OLAINC Ext. 700 TF - 844-465-2462 Ext. 700

[themaddoctor]

Yes, I am lurking.