Formatting an external drive using different interfaces

[jaclaz]

Is that what you would expect?

 

Well, no. , what is strange is that it seems that there is a sort of "transparent UAC" in the latter

 

On windows XP there is no difference between an eSATA and a USB connected device and the dsfi tool works "normally" on both.

 

From what you report it seems like the stupid Windows 8.x makes a difference (BTW, though I believe that Windows 8.x is very stupid , I didn't thought that it could be this much stupid )

 

The "Access is denied." when the command is run with the eSATA connected disk is IMHO "normal", you are running the command in a NON elevated command prompt, the stupid UAC intervenes and denies access, this is reported as an ERRORLEVEL of 1 (the command failed).

 

When you run the same command, in the same NON elevated command prompt, the command apparently goes through, i.e. the correct message from dsfi is displayed and the ERRORLEVEL is at 0 level - success - BUT the target disk is NOT changed

 

I can understand how Windows (including XP) may want to distinguish from an "internal" disk (AFAIK an eSATA connected disk is NOT different in any way from a SATA connected disk inside the PC case) and an "external" disk (please read as USB connected) see also:

http://reboot.pro/topic/9461-page-file-in-usb-hard-disk/

 

But I would have expected that the UAC either protects a disk (if "internal" i.e. IDE/SATA/eSATA) or doesn't protect it (if external, i.e. USB), not that how it seems in this latter case it lets the tool write "virtually" (possibly to a cache of some kind and then fails in flushing the cahce to the device).

 

BTW it is not a real issue , in practice we can go back to executing anyway the tool through execute.exe (and thus provoking the unneeded prompt in XP ) or at this point maybe better add to the batch a small OS version detection routine and let it decide, depending on the OS version detected whether to run the command through execute.exe or directly.

 

@dencorso

Thanks.

I know that I am old-fashioned, but I chose this elevate.exe:

http://code.kliu.org/misc/elevate/

over (say) this one:

http://jpassing.com/2007/12/08/launch-elevated-processes-from-the-command-line/

because it is 5 kb as opposed to 69 Kb , I fail to see the point in having people downloading and installing 432 KB - 10.9 MB* (or more) of MS bloat

 

 

jaclaz

[dencorso]

@dencorso

Thanks.

I know that I am old-fashioned, but I chose this elevate.exe:

http://code.kliu.org/misc/elevate/

over <snip> of MS bloat

 

Sure, but my links were regarding obviating the UAC prompt, which is not avoided by "elevate.exe"... so I guess you've missed my point.

[jaclaz]

Sure , but the point is m00t (with all due respect).

 

It is all about fairness , IMHO people using 8.x do deserve to be nagged by the UAC , the point was only about the current simple implementation that adds the UAC prompt to the poor, innocent XP users, and about having a simple way to have the thingy works as it should (UAC prompt on OS that woud normally require it, NO UAC prompt on OS that would not normally require it) without implementing a Windows OS version detection routine, i.e. finding a simpler solution for that.

 

The queer thing is the behaviour of the command on the USB connected disk, as said before.

 

jaclaz

[Dave-H]

Sorry if this is something obvious that I should have mentioned earlier, but could the different behaviour on the two machines be anything to do with the fact that the eSATA connected 8.1 is 64 bit and the USB connected 8.1 is 32 bit?

[bphlpt]

To throw another question at you, do you have the eSATA and USB links connected at the same time?  I have had problems in the past, for example, when I connected a printer that had two different interfaces to two different computers at the same time, so I thought that there is a possibility that this situation might be similar.

 

Cheers and Regards

[Dave-H]

Hi  bphlpt.

Both interfaces are physically connected at the same time, but there is a switch on the enclosure to select one or the other, and I've been assuming (perhaps wrongly!) that when one is selected the other is isolated.

I've not seen any behaviour so far to contradict that.

[bphlpt]

Probably not a factor then.

 

Cheers and Regards

[dencorso]

Both interfaces are physically connected at the same time, but there is a switch on the enclosure to select one or the other, and I've been assuming (perhaps wrongly!) that when one is selected the other is isolated.

I've not seen any behaviour so far to contradict that.

 

Even so, it's quite easy (and harmless) to disconnect the USB cable, so please do test it. If the behavior is the same with USB either connected or disconnected, when the switch indicates USB is turned off, you may skip doing the reverse test on e-SATA.

[Dave-H]

Just to update, I received my "new" WD1002FAEX drive today, and I've put it in the enclosure to test it.

 

I was quite taken aback to find that it still contained the entire contents of someone else's computer, including all their pictures, videos, and music files!

 

I actually think that's right out or order, selling on a hard drive with someone else's data still on it without re-formatting it, or at least deleting the partitions.

I will be contacting the person i bought it from to tell them this.

There was nothing sinister on it I'm glad to say, but even so I think this it's wrong on principle!

 

The drive seems to work fine, and as expected/hoped, it works as a normal drive identically using both interfaces with both computers.

I'm quite happy to carry on doing the testing with the Advanced Format drive, so do feel free to carry on with this jaclaz!

[NoelC]

That sounds like a surprise, but perhaps not all that surprising.  Why would someone be selling a perfectly good and relatively modern drive?  Computer failure, maybe.  No way to clean the drive.

 

Or it might have been stolen.

 

Not sure if you've already cleared all the files off it, but maybe you could check to see if the person you bought it from is the one whose files you found.

 

-Noel

[Dave-H]

I don't think that the person I bought the drive from on eBay is the person who it originally belonged to.

How the drive got into the seller's hands I have no idea, but there seems to be no connection.

[jaclaz]

 

I'm quite happy to carry on doing the testing with the Advanced Format drive, so do feel free to carry on with this jaclaz!

Good

I am still thinking about problems that this overlapping may cause and evaluating simplicity against "features"/limitations.

I have put together a few working schemes, one with both the FAT12 and the NTFS volumes inside extended, one with the FAT 12 primary and the NTFS volume inside extended (on both 512 and 4096 geometries), one where the FAT12 is "fixed", primary and the NTFS one is also a primary and it's MBR entry is rewritten at each switch (more "classical")

The first one, on one hand is more elegant as it occupies only one entry in the MBR and this entry in the MBR needs not to be modified once the "base image" is deployed, the second is "better" though it will "waste" more bytes (unusable) for the FAT12, BUT both will have a mismatch in Extended partition size, the third has the disadvantage of requiring to correct both the MBR and the VBR of the NTFS partition at each switch (something - rewriting the MBR partition table at each switch - that I would prefer to avoid, and it has to be seen if doable at all if the switcher.cmd is run from within the FAT12 volume).

So, possibly the "right" solution is the second one, the limits would be that two entries in the MBR are occupied, but it would allow, at least when connected through a 512 bytes/sector interface, to boot from the device, which is IMHO a definite plus.

 

jaclaz

[Dave-H]

Hi again jaclaz.

I've checked out my "new" WD1002FAEX drive, and all seems well with it.

I've used it to back up both my Windows 8.1 systems and I've now put it to one side and re-connected the Advanced Format drive.

 

Firstly I must apologise yet again for being really stupid!

The reason why the drive wasn't being converted using your command line on the USB interface was simply because although I changed the drive letters to be correct for the USB connected machine, I forgot that I also needed to change as512.bss to as4kb.bss on the command line!

It does now work using the appropriate commands on both machines.

It is still only putting up the UAC prompt on the eSATA connected machine though, not the USB connected machine.

 

The other strange difference is that the conversion and detection by Explorer of the new drive is immediate on the USB machine, whereas on the eSATA machine nothing seems to happen once the UAC prompt is dismissed until I actually open Explorer At that point, but not before, the new drive appears.

[jaclaz]

So, if you change the relevant lines in the batch:
 

SET Confirm=NOSET /P Confirm=You need to type YES to confirm:IF "YES"=="%Confirm%" dsfi \\.\%SecondDriveLetter% 0 0 %Source% &GOTO :CheckGOTO :EOF

 
as follows:

SET Confirm=NOSET /P Confirm=You need to type YES to confirm:IF /I "YES"=="%Confirm%" (dsfi \\.\%SecondDriveLetter% 0 0 %Source% || elevate.exe -c -w dsfi \\.\%SecondDriveLetter% 0 0 %Source%GOTO :Check)GOTO :EOF

The switcher.cmd should work in all CaSeS (if you pardon me the pun ):

• on Windows 8.x (or more generally in Vista  and later) prompting for UAC when needed 
• on Windows XP not prompting for UAC
• and YES and yes would do as well

jaclaz

[bphlpt]

@jaclaz and @Dave, once you both are satisfied with this "new" tool, could one of you please post the final script and all other necessary components, such as the revised elevate and dsfi or links to them, etc, and an appropriate ReadMe in one place to make it easier to refer others to who might have this same problem?  Thanks in advance.

 

Cheers and Regards

Edited January 16, 2015 by bphlpt