MHDD Usage Questions

[maximus]

This is not "how do I use MHDD", but instead "how do YOU use MHDD". I want to know what functions and features you, as the end user, use of MHDD.

I am making major improvements to my HDDSCRIPTTOOL project for Linux, and I would like to know what should be included for the next release. Please feel free to respond with what you find most useful about MHDD, so that I may try to include those functions in my next HDDSCRIPTTOOL release. When doing so, also understand that I have only used MHDD a few times for very basic things, so I am not fully familiar with all its functions.

[labtech]

Reading the MHDD manual could provide a lot of useful info

[maximus]

Reading the MHDD manual could provide a lot of useful info

It has been awhile since I looked at the manual, but I seem to remember that it was a bit lacking in information. Anyway, what I am trying to ask it what features does everybody use the most, and why? This input will be used to determine what abilities I first put into my software release. I can (and will) add more later. I am not trying to clone MHDD, but instead create new software with growing abilities. Therefore I do not intend to go through all MHDD functions and recreate them, unless they are currently useful.

[labtech]

Firmware accessibility on drives would be cool within a tool that is Linux based.

[Spildit]

Firmware accessibility on drives would be cool within a tool that is Linux based.

Of course that would require VSC for each model that the tool would have support to ...

[maximus]

Firmware accessibility on drives would be cool within a tool that is Linux based.

Of course that would require VSC for each model that the tool would have support to ...

Yes, firmware access would require Vendor Specific Commands. Most of those commands are not publicly available. Therefore I do not have access to them. At this time, hddscripttool will only include things from the ATA standard, and perhaps some publicly well known VSCs. For instance, extracting the password from WD drives is very public (there is a huge thread on here about it). So that is something that I could include (and did include the WD module dump scripts in the first release).

One nice thing about hddscripttool is that since it is script based, with a little bit of effort to learn the language someone could write their own scripts to do what they wanted, including VSCs (assuming that person knew the VSCs and how to use them).

[Spildit]

Any luck compiling some VSC ? It would be a good thing if you could offer a paid version of your software with VSC included, even if they were to be "encrypted" to be used with paid versions of hddscripttool ...

In any case options that are often used with MHDD is the ID/EID, Scan the drive (verify) and lock/unlock with ATA password, relocation of sectors (either erase or remap bad blocks), check S.M.A.R.T., etc ...

[maximus]

I have not yet tried finding other VSCs. I am focusing more on getting the basic functionality working the way I want it to. I am at the point where I am starting to rewrite the scripts. As I do so, I keep finding things I don't like and changing them. The new version will be capable of being text menu driven, with user input so that variables will not have to be set on the command line. It can even list all available drives with model and serial number to choose from.

I just realized I still have to make a script to get smart data and display it. That may be important...

I have a good question. How does the scan with remap work? I know how the scan is performed using the read verify command, but is remapping simply done by writing to the bad sector? The MHDD documentation states:

Then you have to run ERASE command which erases every sector on your drive, and drive recalculates ECC fields of each sector. This helps to get rid of «software bad blocks». If that do not help, you have to run SCAN with REMAP option switched on.

So what does the erase command do if it doesn't remap? I would think that erase would be writing 0's to all the sectors, and if the sector was bad the drive should remap it.

Also, what is the difference between low level formatting and writing 0's to every sector on the drive? Is there a difference?

[maximus]

And I am thinking the scan function could use some improvement. I am thinking it could write the actual time of each read to a text file. Column 1 would be the starting sector number, column 2 would be the time, and column 3 could be the error if any. The file could get a bit large on big drives, so with 48 bit the read size could be increased from 256 sector to something bigger up to 65536. The file could then be plotted using something like gnuplot so speed changes could be seen visually. Would anyone be interested in that?

[fzabkar]

Also, what is the difference between low level formatting and writing 0's to every sector on the drive? Is there a difference?

LLF is a misnomer. Low level formatting hasn't been possible since the introduction of IDE drives way back in the early 1990s.