Export thread

With this release, ALL KNOWN PROBLEMS with SpinRite 6.1, including its operation on very troubled drives, are resolved and have been confirmed by everyone who has experienced them. This update:

• Fixes the temperature logging bug I introduced in release candidate 4, which added logging of drive over-temperature events.
• Resolves a problem Paul Farrer discovered with USB BIOS not supporting a status checking command SpinRite was using.
• Adds a final command-line option “skipverify” which forces SpinRite to bypass all of its drive access confidence and quick benchmarking tests. This was added after someone here in the forums encountered a drive that Windows could access, but which was so fragile that it was pushed into Device Fault by SpinRite's early probing. Since some of SpinRite's initial confidence testing work occurs at the very end of the drive, forcing that work to be skipped might allow SpinRite to be used across most of such drives rather than never being able to even run on such a drive. (Note, however, that any such drive should never be kept in service beyond pulling it data off.)

At this point, unless and until new problems arise, I'll be working on an improved “remaining time to go” estimator to replace SpinRite's current linear extrapolation. Linear extrapolation underestimates remaining time on all modern zone-bit recorded spinning drives whose performance decreases as the scan progresses. So SpinRite will be getting an improved estimator which will take this gradual decline in performance into account.

As always, you may obtain this update by going to GRC's Pre-Release page, entering your SpinRite license serial number or transaction code and clicking the links to obtain your personal pre-release copies of either the full Windows/DOS SpinRite and/or just the DOS code.

(VirusTotal is 100% happy with it, giving it 0/67 detections.)

Thanks, all!

Excellent news!

Hello Steve,
After listening to you say in SN956 that SpinRite 6.1 doesn't run on virtual machines, I decided to try it myself because my use case differs from the one you reported. I found that it's possible to run on Virtualbox 7.0 with some limitations. I am writing about all the processes I went through, hoping it might be helpful to someone else.

I use Linux as my main OS, but occasionally, I have to use Windows, and for that, I use Virtualbox 7.0. So, I've downloaded SpinRite 6.1 PR 5.06 into my Windows VM. First problem: Windows Defender flagged SpinRite as malware:


Because each customer has their own signed SpinRite version, everyone has a different file, so I decided to check it also on Virustotal. Good news, I got zero/65, as you mentioned in SN

However, this doesn't solve the problem of my Windows flagging SpinRite as malware and deleting it immediately. So I had to add one folder as a "security exception" (https://answers.microsoft.com/en-us...defender/8a096ae3-5ee5-4eac-9dc7-ab4d8628b035) and redownload SpinRite PR. Now, I could finally run SpinRite and create the ISO file.

The next steps are done on my Linux system (Ubuntu 22.04). I connected my SSD to a USB port and created a virtual disk (VMDK) that pointed to my physical device with this command:

sudo vboxmanage createmedium disk --filename ~/ssd.vmdk --format=VMDK --variant RawDisk --property RawDrive=/dev/sda

Note: This command is specific to Virtualbox version 7. For previous versions, the command syntax is different.

Next, on my Linux:

1. I opened Virtualbox with sudo (without sudo, it doesn't work)
2. created a new DOS VM, and added the SpinRite ISO file as a CD with boot device and the SSD.vmdk as a disk
3. I booted the DOS VM, and SpinRite worked without a problem (see images below). Note: the SSD I used for this test has no known issues.

Limitations of this solution:

• Because the ISO file is read-only, SpinRite cannot create log files. We can solve this by adding another virtual disk to the Windows VM, but this time as a USB device. Then, use Rufus to create a USB bootable device from the SpinRite ISO file. Use this USB virtual disk in the DOS VM instead of the SpinRite ISO file as your boot disk.
• The other limitation is that there's no Smart data. I'm unsure if this is a USB or VM limitation. I also don't know if this affects SpinRite's ability to recover storage devices with problems. Maybe you can enlighten us about this.

VERY nice posting and lovely screenshots, @digicromo.

digicromo said:

I found that it's possible to run on Virtualbox 7.0 with some limitations.

Click to expand...

<<grin>> I actually did a great deal of development of SpinRite v6.1 under VirtualBox. It's my favored VM environment and there's some explicit support for it in v6.1. I needed to tweak SpinRite's native AHCI driver since VB's AHCI adapter simulation acts slightly different from real hardware. But adding any sort of “layer” between SpinRite and the drive introduces a degree of uncertainty. So... GRC's “official” position needs to be that use under VM's is not supported.

And your screenshot(s) above demonstrate a perfect example of this: The second screenshot shows that SpinRite is using direct bus mastering IO with that drive, That means that it should have access to the drive's SMART data... and it definitely would if it were being run outside of a VM directly on the machine's hardware. So you're also correct to note the limitations that this imposes.

Great posting, @digicromo. Thanks!

@Steve, thank you for your quick reply and explanation.

I'm sorry if my post made you think I was implying that you should give official support to run SpinRite inside a VM, that wasn't my intention. I think that using a VM is an excellent (and cool) way to run SpinRite when we don't have a spare computer to run SpinRite exclusively or for a live software demo in the middle of a presentation.

Occasionally, I give digital forensic training to LEA in my country. Once, I mentioned your software to help solve the slowness in old HDDs that needed to be forensically copied. Initially, no one was impressed with my description of the tool, but then I made a live demo using Virtualbox and caught everyone's attention. I guess people need to see it to believe it. Because both physical space and computers are scarce resources for LEA one of the trainees asked if it was possible to run SpinRite in multiple drives at the same time using Virtualbox. I never tested that possibility, but now that I remember this story, I'm eager to try it.

digicromo said:

Initially, no one was impressed with my description of the tool, but then I made a live demo using Virtualbox and caught everyone's attention.

Click to expand...

I agree with and understand the power of such demos. Once I get to the documentation phase I plan to produce a couple of videos since that has become everyone's preferred medium.

So long as you only do a non writing scan, it should be fine.

Hello, Steve, thanks for this masterpiece of software! My Issue:
Ventoy-Compatibility ist broken in Spinrite 6.1.
The Spinrite 6.0 Floppy Image loads fine (with MEMDISK on), 6.1 does not.
Can you fix this? Or is it a Ventoy issue?

@Juppi:

I recently initiated some discussion of this over in GRC's newsgroup. I haven't yet given it my full attention since I recently decided that SpinRite needed to offer a better solution to the bug we discovered in the USB portion of many AMI system BIOSes. So I'm going to get that resolved then look into Ventoy compatibility since it's clear that many of SpinRite's power users want this capability.

I've never used Ventoy myself. So that will need to change. <g> But without having looked into what's going on, no reason stands out to me for why v6.0 would work whereas v6.1 does not. A big question in my mind is about the idea that ISO media is generally treated as read only and SpinRite wants to save its .CFG settings file in the same directory from which it was run, and its logs in an SRLOGS sub-directory. But SpinRite also uses the "diskette emulation" mode of ISO booting, so perhaps that explains why it works?

Anyway, I will be looking into this question before v6.1's code is declared final.

Hello
I have followed the instructions to run the pre-release, created a USB-stick that worked fine with 6.0, copied the new 5.06 file and renamed both the old and the new file. But when I boot up I receive the message "This Pre-Release of SpinRite v6.1 has EXPIRED!!"
Am I doing something wrong, I thought it was meant to work until April?

Stybbe said:

EXPIRED

Click to expand...

Is the clock on your PC correct?

PHolder said:

Is the clock on your PC correct?

Click to expand...

Sorry. Why didn't I check that first. The year was 2085....

Thank you!

FWIW, I've downloaded and installed the latest pre-release. Only positive feedback, everything worked perfectly (multiple level 2 and 4 scans). Thank you.

@MikeJc: Thanks Mike. I appreciate the feedback. It's been a long but gratifying three-year journey. But I'm also feeling good about where we are... and just about finished!

Long time "watcher" and "fan" but first time posting.

I downloaded and tried out the current RC and tested it on a few of my systems. I made only one thumb drive so after I started the scan function, I removed the thumb drive for the next system. The scans went fine but I got an error that the drive was not installed for the log file write, which I expected. BUT when I tried to do any more functions, another scan or benchmark, it woudl go to a black screen and nothing else, until I rebooted with the thumb drive reinstalled to reload the app. Not a major issue but since I closed on continue, I woudl think that SR woudl go ahead without a log. I also submitted a "feedback" on the GRC.com web site since I was not able to post until now.

Thanks, @cemory,

As you noted, everything you described is (unfortunately) expected. The original BIOS has grown quite old and (now) creaky. USB came along well after it was established. So the BIOS has never had any provision for managing the "on the fly" "plug-n-play" concept of USB devices being inserted and removed. It can handle that for diskettes, but even there, the drive remains connected while the media may change.

And all of this is why the moment SpinRite v6.1 is wrapped up, I'll begin work on a new SpinRite that will be leaving the BIOS behind... and that SpinRite (7+) will be able to handle dynamic USB drive insertion and removal. <g>

Thanks again for your testing and feedback!

Hello,

Steve, thank you for the great software!

I'm just dropping in to share a quick observation.

I tried running SpinRite via PXE boot (boot from LAN) with MEMDISK emulating the floppy disk.
DOS boots up fine and shows the SpinRite splash screen, but then it just goes to a blank screen with a blinking cursor. No further action from there.
Notes:
- SpinRite 6.0 floppy image works with the same setup (PXE);
- SpinRite 6.1 works from USB floppy drive on same PC (physical floppy from same image))

This looks similar to Juppi's issue with Ventoy, considering both use MEMDISK (part of Syslinux).
Also, I checked, and the floppy is emulated in read-write mode, so I don't think that read-only ISO is to blame.

Since this appears to be a "known bug," I just wanted to add more information and context in the hopes that it might assist in resolving the issue.

Hello,

I'd heard mention on the Security Now podcast that the final release of SR6.1 will not have the 137GB limit on USB drives? I have a couple of WD passport 2.5" drives that I cannot scan fully. WD, in their twisted wisdom, installed the USB 3.0 adapter directly to the drive board. I previously thought I could remove the drive from the housing then install into a system as an internal device to scan. Don't know if other manufacturers are doing this nasty trick.

@waltmania:

You heard correctly, Walt. I have it running now and it's working very well. Since I'd like to minimize the rate of updating, I'm working on a few more bits... But I'm certain I'll have something solid be the end of this weekend at the latest. Perhaps today (Saturday). I'll post a notice as I have been at the top of the forums once it's ready.

waltmania said:

I'd heard mention on the Security Now podcast that the final release of SR6.1 will not have the 137GB limit on USB drives?

Click to expand...

Mostly. The limit will now be 2.2 TB.

Many of the USB drives do this, as that saves on cost, and they sell a lot of the drives, so putting this function into the controller, and the appropriate interface, is a saving in cost, removing an expensive connector pair (gold plate is expensive), an extra board with components, and a few mm of space in the outer case.

waltmania said:

Hello,

I'd heard mention on the Security Now podcast that the final release of SR6.1 will not have the 137GB limit on USB drives? I have a couple of WD passport 2.5" drives that I cannot scan fully. WD, in their twisted wisdom, installed the USB 3.0 adapter directly to the drive board. I previously thought I could remove the drive from the housing then install into a system as an internal device to scan. Don't know if other manufacturers are doing this nasty trick.

Click to expand...

I've got a friend with a WD drive exactly like this, and was very pleased to hear that Steve has addressed the BIOS problem that led to him adding this limitation in his earlier 6.1 PR candidates. I look forward to being able to run the full scan on this drive (and, here's hoping that the lack of direct SATA access to the drive doesn't impede Spinrite from fixing whatever is wrong with this one).