[joe tweaker]

I have a new AcomData Samba External USB hard drive, which is an enclosure containing a USB interface and a 160GB Western Digital PATA/IDE drive. To ensure compatibility with my aging ABIT BM6 motherboard and Windows 98se, I am only using the first 128GB of the drive. The drive was partitioned as follows while installed as the Master Drive on the secondary IDE cable, then removed and placed into the AcomData Samba External USB drive enclosure. I created NO primary partitions as this drive is intended for data storage only. It will never be used as a boot drive. I created ONE extended partition of 128GB (to avoid data corruption) and FOUR logical partions of 31.8GB within the extended partition. The rest of the drive is not used.

Sometime in 2007 I used the Soporific Autopatcher to update 98SE to the latest patches. One of those was the Maximus-Decim Native USB drivers, but I have no idea what VERSION of those I have installed now. They have worked great with a 128MB Memorex Thumb Drive in the two USB ports built into the rear of my ABIT BM6 motherboard. Those ports are either USB 1.0 or 1.1, NOT USB 2.0. The BM6 owners manual only refers to them as "USB" ports, so they may only be 1.0. This doesn't appear to be a problem with either the Memorex Thumb Drive OR the new AcomData Samba / 160GB Western Digital combo. When I attach the AcomData to either USB port on the BM6 motherboard, the NUSB icon appears in the systray, installs each of the four logical partitions, and reads and writes data perfectly at whatever speed it is capable of.

The problem is if I forget to STOP the external drive and POWER IT DOWN before every reboot, it is causing my system to crash with a "Fatal Exception 0E" at 0028: FF0328FC in VXD USBD(05) +0000169C when it reboots to windows. It does not appear to be losing any data when this occurs. I merely have to switch it off, reboot again, then switch it back on AFTER the windows desktop appears. I have saved a BOOTLOG.TXT during one of these restarts to see where it fails, and it appears to occur AFTER processing both CONFIG.SYS and AUTOEXEC.BAT, and AFTER running my startup AVG virus scan, but BEFORE reaching the windows desktop. It crashes consistently at the same point anytime I forget to switch it OFF before doing a restart.

My question: Could this have something to do with it not having a primary partition? I have internal drives partitioned the same way (so the drive letters are always assigned in the same order), and they never pose any such problems during startup. Could it be due to my motherboard's USB ports being older than USB 1.1? Again, if I don't switch the drive ON until after the windows desktop appears, the drive seems to work perfectly from then on. Could it have something to do with my using an older version of Maximus-Decim's NUSB driver? (I have no idea which version I have installed.) Or could it be an IRQ conflict? In the Device Manager it shows IRQ 11 is being shared by both the Intel 82371AB/EB PCI to USB Universal Host Controller AND by my Creative Labs Sound Blaster Live! Value sound card, but it also shows an entry for IRQ Holder for PCI Steering on IRQ 11 as well. Again, it has no problem reading or writing to the external hard drive IF the drive is not turned ON until after windows has reached the desktop. The Fatal Exception 0E crash / lockup ONLY occurs if the drive is NOT powered off when I do a system restart.

I have a friend who has an external 500GB WD Book Drive and windows XP, and she can leave it powered on throughout every system restart she does. When I attach my AcomData enclosure to her system, it stops her system cold just after it displays the video card logo when I reboot with it powered ON. eg, it doesn't even enumerate other devices until my AcomData is turned OFF. As soon as it's switched off, however, her XP system resumes without any help whatsoever. On mine, it appears to have no problem enumerating, and completing most of the boot process, BEFORE the Fatal Exception 0E appears, and then I'm "dead in the water." (I have to switch the AcomData enclosure OFF, AND press the "reset" button to get mine to continue.)

FWIW, when I first rec'd this drive (as a gift from a friend) it exhibited the same behavior if powered on at startup, but my friend had assembled it with the WD hard drive jumpered to "Cable Select." The owners manual clearly states it must be jumpered as a "Master" so I took care of that when I had it hooked to the internal IDE cable and repartitioned it to avoid exceeding the 128GB limit with 98se. Since then it has been jumpered as "Master" but this appears to have had no impact on the startup issue.

Anyone else ever face and resolve this issue? Or have any idea why it has to be OFF during a restart, but a WD "Book Drive" doesn't? I really would like to get this fixed so I can leave the drive ON during system restarts!

This post has been edited by joe tweaker: 20 March 2009 - 04:19 AM

[joe tweaker]

Some additional info:

I checked the BIOS to be sure there was no option set to boot to a USB drive.
No such option exists. I tried A,C,SCSI and C,A,SCSI and nothing changed.
Support for USB keyboard is set to OS. I have a standard PS2 keyboard.
There is no BIOS setting for "Legacy USB support" (The last BIOS was written before USB 2.0).

The last entry in the aborted BOOTLOG.TXT says "[00028E20] Dynamic load success lmouhid.vxd"
The previous 12 lines appear to be loading drivers for my Logitech Mouseman Cordless PS/2 mouse.
If I hunt for the next line after that in a sucessful bootlog (external USB OFF during restart),
it says "Starting Logitech-compatible Mouse (PS/2) (BIOS\*PNP0F13\0A).

Does this mean my cordless PS/2 mouse has a conflict with my external USB drive
when it's ON during a system restart? But it's not in conflict once the desktop is loaded?
And how could this tie in with the freeze at the video card logo on the XP system?

Ideas? Anyone?

[risk_reversal]

Quote

joe tweaker said
I checked the BIOS to be sure there was no option set to boot to a USB drive.
No such option exists. I tried A,C,SCSI and C,A,SCSI and nothing changed.
Support for USB keyboard is set to OS. I have a standard PS2 keyboard.
There is no BIOS setting for "Legacy USB support"

In the first instance, have you tried setting Support to USB Keyboard to Disabled.

I must admit I never allow my external hdd to be connected to my pc at boot (if they merely contain data). Much prefer to just connect it afterwards. Avoids any possible problems and much safer in my opinion.

Quote

joe tweaker said
To ensure compatibility with my aging ABIT BM6 motherboard and Windows 98se, I am only using the first 128GB of the drive

External usb enclosures are not subject to this barrier.

Quote

joe tweaker said
I created NO primary partitions as this drive is intended for data storage only. It will never be used as a boot drive. I created ONE extended partition of 128GB (to avoid data corruption) and FOUR logical partions of 31.8GB within the extended partition. The rest of the drive is not used.

Hdds are allowed up to 4 primary partitions. One of those primary partitions can be an extended partition containing a number of logical partitions.

Creating an extended partition on the external hdd allows for greater flexibility ie the set up that you have. But you could just as easily created 4 primary partitions.

Quote

joe tweaker said
One of those was the Maximus-Decim Native USB drivers, but I have no idea what VERSION of those I have installed now

NUSB v2.x installs just the usb drivers and works with usb1.x and usb2.0. NUSB version 3.x add usb2.0 drivers also but you have to uninstall your usb2 controller driver prior to it's install. I have usb2.0 on my mobo(s) but use NUBS v2.x as the mobo's usb2.0 controller is working fine with no issues.

Quote

joe tweaker said
The problem is if I forget to STOP the external drive and POWER IT DOWN before every reboot, it is causing my system to crash with a "Fatal Exception 0E" at 0028: FF0328FC in VXD USBD(05) +0000169C when I reboot

Can't see why it is a problem to just disconnect via the NUBS systray icon. I also occassionally forget and sometimes get a BSOD. Does this have anything to do with the fact that the external HDD has an extended partition rather than a primary one. I very much doubt that.

Quote

joe tweaker said
I have a friend who has an external 500GB WD Book Drive and windows XP, and she can leave it powered on throughout every system restart she does. When I attach my AcomData enclosure to her system, it stops her system cold just after it displays the video card logo when I reboot with it powered ON. eg, it doesn't even enumerate other devices until my AcomData is turned OFF. As soon as it's switched off, however, her XP system resumes without any help whatsoever.

I will have a stab at this one. WD retail external enclosures have proprietary data back up software. If your friend also has system restore turned on, I am guessing that the freeze is due to XP trying to reconcile the contents of the system information folder (system restore) from the WD with your drive and it is getting confused. If I have understood your question correctly. Bottom line, safer to have the external hdd disconnected at boot. Your friend's set up works but the potential for errors to occur is in my opinion great.

Good Luck

This post has been edited by risk_reversal: 18 March 2009 - 05:10 AM

[joe tweaker]

First, let me thank you, risk_reversal, for being the first to respond!

Quote

risk_reversal said
Have you tried setting Support for USB Keyboard to Disabled.

The only choices are BIOS or OS. It is currently set to OS. Disabled is not a valid choice.
The BM6 owners manual says choosing BIOS turns ON support for a USB keyboard via the BIOS.
So I assumed choosing OS was equivalent to DISABLING said support via the BIOS.
I use a PS/2 keyboard. Would you leave this setting at OS? Or change it to BIOS?

Quote

risk_reversal said

Quote

joe tweaker said
To ensure compatibility with my aging ABIT BM6 motherboard and Windows 98se, I am only using the first 128GB of the drive

External usb enclosures are not subject to this barrier.

Are you certain about this? How is that accomplished? Are addresses above 128GB handled by a second BIOS in the USB enclosure instead of the BIOS on the motherboard? Should each partition still be limited to 128GB when using 98se as the O/S?

FWIW, when I created the current partitions, the drive was not in the USB enclosure. It was on the internal IDE cable, and I used Bootit-NG from Terabyte Unlimited, which includes a partitioning tool, to create and format the current partitions. Bootit-NG can also work with hard drives that are connected to a USB 2.0 port, but the USB ports on my aging BM6 motherboard are 1.x, not 2.0. Thus I had to do the partitioning on this drive via the internal IDE cable, and apparently that limited me to the first 128GB of the drive.

How exactly would I repartition within the USB enclosure to use the rest of the drive? Especially when the USB enclosure itself CANNOT BE TURNED ON under 98se (or XP) until AFTER the windows desktop is loaded? What partitioning tool works with USB drives from WITHIN a windows environment?

Obviously, if going beyond 128GB is possible ONLY when the drive is within the USB enclosure, I would never be able to remove that drive and connect it directly to the intenal IDE cable again without risk of data corruption, but perhaps I could live with that limitation in exchange for the extra storage.

Quote

risk_reversal said
Hdds are allowed up to 4 primary partitions. One of those primary partitions can be an extended partition containing a number of logical partitions.

Creating an extended partition on the external hdd allows for greater flexibility, ie the set up that you have. But you could just as easily created 4 primary partitions.

Does this (BSOD) have anything to do with the fact that the external HDD has an extended partition rather than a primary one? I very much doubt that.

Having built my own systems for more than a decade, and having used Partition Magic and Bootit-NG for years, I know very well that creating up to four primary partitions is possible. But I also hate having the O/S rearrange my drive letters whenever one or more of my hard drives is swapped or removed from my systems. Placing all logical drives in extended partitions EXCEPT for the BOOT partitions on the first internal drive makes it much easier to manage the drive letter assignments. I've just never employed this strategy on a USB hard drive before.

Quote

risk_reversal said
NUSB v2.x installs just the usb drivers and works with usb1.x and usb2.0. NUSB version 3.x add usb2.0 drivers also but you have to uninstall your usb2 controller driver prior to it's install.

So how do I uninstall my older version of NUSB before installing the newer one? The older one (which was installed using the Soporific AutoPatcher in 2007) isn't listed in the Add/Remove applet in the Control Panel. Would I open the Device Manager and remove all the USB Root Hubs and then the Intel USB Host Controller? If I do that, won't it just reinstall the existing older driver on the next reboot?

Quote

risk_reversal said

Quote

joe tweaker said
The problem is if I forget to STOP the external drive and POWER IT DOWN before every reboot, it is causing my system to crash with a "Fatal Exception 0E" at 0028: FF0328FC in VXD USBD(05) +0000169C when I reboot

Can't see why it is a problem to just disconnect via the NUBS systray icon. I also occassionally forget and sometimes get a BSOD.

It's a problem because it's not always possible to disconnect via the systray icon before a restart. ie, When the power suddenly goes out, or the system has frozen or locked up. And, when I'm in a hurry, I have a tendency to forget and just reach for the reset button. In situations like these, it would be much less aggravating if my system would just restart without these damned BSODs.

You didn't comment on the shared IRQ situation. Thus I'm still wondering if either that, or the Logitech Cordless Mouse, could be causing the BSOD when I restart my system (but ONLY when the external USB drive is ON)?

I failed to mention earlier that in addition to her Western Digital Book Drive, my friend also has a SimpleTech external [re]drive attached to her XP system. Neither of them have to be turned off when she reboots, and I don't think she installed any proprietary software that came with either of them. She just plugged them in, and they stay ON all the time. Without any need to be powered OFF when she restarts or reboots her system.

This is only a guess, but I suspect her system halts earlier than mine, and recovers easier when my drive is attached, because her motherboard has a newer BIOS than mine, and has USB 2.0 ports, whereas my older motherboard only supports USB 1.x. I'm willing to live with the slower transfer rate of 1.x, but still puzzled by the fact this AcomData drive has restart issues on BOTH systems, not just on mine. ie, By failing on more than one system, even in slightly different ways, it looks more like the problem is in the drive itself, and not in my BIOS. Perhaps this is just an illusion, but without feedback from others, I have no way of knowing.

I'd still like to restart my system without having to turn this drive OFF, then ON again EVERY TIME. But without more input, I'm drawing a blank on how to proceed.

[risk_reversal]

Quote

joe tweaker said
Are you certain about this? How is that accomplished? Are addresses above 128GB handled by a second BIOS in the USB enclosure instead of the BIOS on the motherboard? Should each partition still be limited to 128GB when using 98se as the O/S?

About a year ago I searched this very topic on this forum. The many posts to be found confirmed this. I purchased an external enclosure with a 250gb HDD and have had not probs. The HDD has a single partition has about 90gb left (so I am well over the 128GiB limit).

Quote

joe tweaker said
FWIW, when I created the current partitions, the drive was not in the USB enclosure. It was on the internal IDE cable, and I used Bootit-NG from Terabyte Unlimited, which includes a partitioning tool, to create and format the current partitions. Bootit-NG can also work with hard drives that are connected to a USB 2.0 port, but the USB ports on my aging BM6 motherboard are 1.x, not 2.0. Thus I had to do the partitioning on this drive via the internal IDE cable, and apparently that limited me to the first 128GB of the drive.

In which case, then you would have had a 128GiB capacity limitation imposed but the mobos IDE controller. I guess that you can probably connect your external to a mobo with usb2.0 and use BiNG to extend the extended partition to occupy the whole amount. I use BiNG but have never tried using it with an external on usb 1.x.

Quote

joe tweaker said
How exactly would I repartition within the USB enclosure to use the rest of the drive? Especially when the USB enclosure itself CANNOT BE TURNED ON under 98se (or XP) until AFTER the windows desktop is loaded? What partitioning tool works with USB drives from WITHIN a windows environment?[

Boot with BiNG with the external disconnected and after BiNG has loaded connect the external to a usb port. Failing that you could boot with a Puppy Linux Live CD and use GParted (partition utility). After Puppy has booted and you connect the usb drive Puppy will detect it. I have an old Dell with usb1.x and Puppy finds the external.

Quote

joe tweaker said
Obviously, if going beyond 128GB is possible ONLY when the drive is within the USB enclosure, I would never be able to remove that drive and connect it directly to the intenal IDE cable again without risk of data corruption, but perhaps I could live with that limitation in exchange for the extra storage.

Many ways to get around the 128GiB limitations which are listed on this forum. Personally, I use 2 250Gb sata drives in a dual boot system with 98SE/XP. I overcome this by using the Promise controller which loads it's own drivers.

Quote

joe tweaker said
Having built my own systems for more than a decade, and having used Partition Magic and Bootit-NG for years, I know very well that creating up to four primary partitions is possible. But I also hate having the O/S rearrange my drive letters whenever one or more of my hard drives is swapped or removed from my systems

Could always just to into Device Manger and fix the drive lettering for the external...

Quote

joe tweaker said
So how do I uninstall my older version of NUSB before installing the newer one? The older one (which was installed using the Soporific AutoPatcher in 2007) isn't listed in the Add/Remove applet in the Control Panel. Would I open the Device Manager and remove all the USB Root Hubs and then the Intel USB Host Controller? If I do that, won't it just reinstall the existing older driver on the next reboot?

My clear understanding is that there is no uninstall for NUSB. But I guess that if you deleted all the items listed under Universal serial bus controllers in Device Manager (inc usbstor.inf), this may work. You will need to post this as a new question. I would be interested in the replies.

Quote

joe tweaker said
You didn't comment on the shared IRQ situation. Thus I'm still wondering if either that, or the Logitech Cordless Mouse, could be causing the BSOD when I restart my system (but ONLY when the external USB drive is ON)?

Mutiple devices on the usb bus can at times become problematic. Cordless usb mice for me are a nono. I use ps/2 and that will not change anytime soon. Are you also using a powered usb hub. This may help.
In respect of usb clashes, my understanding is that if you have 10 usb devices, each is not assigned a separate IRQ, it is the usb bus that has that function, if you see what I mean. Now your usb controller may be clashing with something else eg I had a NIC which would lock up when using my usb printer. Move the card problem solved. But you do need to know the IRQ shares.

Quote

joe tweaker said
This is only a guess, but I suspect her system halts earlier than mine, and recovers easier when my drive is attached, because her motherboard has a newer BIOS than mine, and has USB 2.0 ports, whereas my older motherboard only supports USB 1.x. I'm willing to live with the slower transfer rate of 1.x, but still puzzled by the fact this AcomData drive has restart issues on BOTH systems, not just on mine

A guess, maybe the usb enclosures's chipset has issues.....I don't think it's to do with the 4 logical partitions.

Well personally I have been having a fun time figuring out how to get a wireless NIC to operate with WPA encryption on 98se. Have now solved that issue but was a bit tedious.

Good Luck

PS

Quote

joe tweaker said
The only choices are BIOS or OS. It is currently set to OS. Disabled is not a valid choice.
The BM6 owners manual says choosing BIOS turns ON support for a USB keyboard via the BIOS.
So I assumed choosing OS was equivalent to DISABLING said support via the BIOS.
I use a PS/2 keyboard. Would you leave this setting at OS? Or change it to BIOS?

Not sure. If you have a usb wireless mouse should this not be enabled? If you want to try setting to BIOS but feel edgy, why not disconnect your hdds and try it at least there will be no harm done. You can just boot to a dos floppy or maybe a Puppy Linux Live CD.

This post has been edited by risk_reversal: 18 March 2009 - 06:17 PM

[Queue]

If this external is making two very different computers choke when they boot up, there's something wrong with that external drive. None of the half-dozen plus externals I've used have ever caused problems if left plugged in through reboots, etc. It could be because of the partitioning scheme, or the hardware itself could be at fault.

Second, splitting up a larger than 137GB drive into multiple 137 or less GB partitions won't protect it on a system that can't work with a drive larger than 137GB; once a physical sector past the 137GB limit is written to, say bye-bye to the data on that drive. Because of the nature of the error that occurs, you usually wind up losing every file (they still exist to some degree, but files are pointing at the wrong data, so data is essentially scrambled). Smaller partitions will make scandisk work better on Win9x (at least) though. The only way to be safe is to limit the drive to 137GB or less of usable space.

Queue

[joe tweaker]

Thanks to both of you for the additional info.

There seems to be some confusion between you about whether or not data corruption can occur on a USB drive beyond the 128GB address barrier (137+ Billion bytes per marketing depts). Risk_Reversal says he has surpassed this limit without any problems, but further down says he has a dual-boot system with a Promise controller installed. I have only a 10 year old Abit BM6 motherboard with two internal IDE channels, and two USB 1.x ports on the backplane. So what worked for him *may* not work for me. If either of you could point me to an existing thread on MSFN where this was previously discussed, I'd be happy to do more research on this.

As for my Logitech Mouseman Cordless Mouse, it's NOT USB, nor does it share an IRQ with the USB Host Controller. It's a cordless PS/2 mouse, and I don't know why it would be the last device successfully loaded in the BOOTLOG.TXT when my system hangs with the Fatal Exception 0E on restart with the AcomData external USB drive ON. It just ends there when the system hangs. It's my Sound Blaster Live! Value sound card that shares IRQ 11 with the Intel82371 USB Host Controller and IRQ for PCI Steering, NOT the Logitech Cordless PS/2 Mouse. So, I have no idea why I'd need to enable USB Keyboard Support in the BIOS for my PS/2 mouse, but I could give it a whirl just to see what happens.

I tried running the latest BiNG from a floppy the last time I was at my friend's house with my AcomData external USB drive. I started BiNG with the AcomData OFF, enabled support for USB 2.0, then turned the AcomData ON. As soon as I turned it ON, BiNG FROZE and all the icons on the screen disappeared. When I turned the AcomData OFF, the icons returned, and I was able to function normally in BiNG again, but my drive was no longer available since it was OFF.

For comparison, I loaded BiNG on my friend's computer with only her Western Digital Book Drive and SimpleTech [re]drive attached. With her two USB drives, BiNG didn't freeze, recognized both drives, displayed their partitions and offered to let me resize, format, or delete them. Because they were not mine and had data on them, I didn't actually test these commands, but they were available and BiNG wasn't locked up as it was when the AcomData was attached and turned ON.

So, after loading the Windows desktop on ANY version of windows with 1.x or 2.0 USB ports, the AcomData works, flawlessly. But turn it ON anywhere else (outside of windows) and everything locks up until it is turned back OFF. On her system, turning it OFF enables one to continue (as if nothing happened), but on MY much older system things stay frozen until the "reset" button is pressed.

I've heard of Puppy Linux, but never worked with it. I might give it a whirl someday when I have more time.

This post has been edited by joe tweaker: 18 March 2009 - 08:02 PM

[dencorso]

There is definitely *NO* 137 GB / 128 GiB limit for USB drives in Win 9x/ME. Only internal IDE (PATA) and SATA (when not using Via's VIASRAID.MPD) present this issue. The problem is related to the driver file ESDI_506.PDR can be solved by using the file patched by LLXX. USB mass storage devices rely, instead, on USBPORT.SYS and USBSTOR.SYS. For more, see the related link in my signature.

[joe tweaker]

dencorso, on Mar 18 2009, 11:15 PM, said:

There is definitely *NO* 137 GB / 128 GiB limit for USB drives in Win 9x/ME. Only internal IDE (PATA) and SATA (when not using Via's VIASRAID.MPD) present this issue. The problem is related to the driver file ESDI_506.PDR can be solved by using the file patched by LLXX. USB mass storage devices rely, instead, on USBPORT.SYS and USBSTOR.SYS. For more, see the related link in my signature.

Thanks for that link, dencorso! Now all I need is some way to repartition that drive from within windows since it locks up everywhere else!!! Any chance you could also provide the correct way to upgrade to the latest Maximus-Decim Native USB drivers from an earlier version? Risk_Reversal said I need to remove the old drivers first, but neither of us is certain how that should be done (since they're NOT listed in the Add/Remove window).

[Queue]

Erm, when I kinda went off on a tangent about the 137GB limit, it was mainly for safety sake in case you ever plug the drive in via an IDE cable and assume things were safe due to having multiple <137GB partitions; a drive running through a USB adapter or PCI card that uses its own drivers (or even a motherboard with its own drivers) as opposed to esdi_506.pdr should be safe. With a patched esdi_506.pdr, any scenario should be safe.

I've had no trouble using a 500GB SATA drive (attached to a SATA PCI card) or a 500GB external USB drive with Win9x with an unpatched esdi_506.pdr.

I'm anxious to hear if reformatting it resolves the issue.

Queue

[Multibooter]

joe tweaker, on Mar 17 2009, 03:32 PM, said:

The problem is if I forget to STOP the external drive and POWER IT DOWN before every reboot, it is causing my system to crash with a "Fatal Exception 0E" at 0028: FF0328FC in VXD USBD(05) +0000169C when I reboot.

If I understand right, you have a shutdown problem when your Acomdata USB drive is connected. Have you tried ScanDisk on your USB drive?

[risk_reversal]

Quote

joe tweaker said:
As for my Logitech Mouseman Cordless Mouse, it's NOT USB, nor does it share an IRQ with the USB Host Controller. It's a cordless PS/2 mouse, and I don't know why it would be the last device successfully loaded in the BOOTLOG.TXT when my system hangs with the Fatal Exception 0E on restart with the AcomData external USB drive ON. It just ends there when the system hangs. It's my Sound Blaster Live! Value sound card that shares IRQ 11 with the Intel82371 USB Host Controller and IRQ for PCI Steering, NOT the Logitech Cordless PS/2 Mouse. So, I have no idea why I'd need to enable USB Keyboard Support in the BIOS for my PS/2 mouse, but I could give it a whirl just to see what happens.

Sorry misunderstood that it was a USB cordless. In which case turning the turning USB Keyboard Support to BIOS would not likely make any difference.

Quote

joe tweaker said:
I tried running the latest BiNG from a floppy the last time I was at my friend's house with my AcomData external USB drive. I started BiNG with the AcomData OFF, enabled support for USB 2.0, then turned the AcomData ON. As soon as I turned it ON, BiNG FROZE and all the icons on the screen disappeared. When I turned the AcomData OFF, the icons returned, and I was able to function normally in BiNG again, but my drive was no longer available since it was OFF.

For comparison, I loaded BiNG on my friend's computer with only her Western Digital Book Drive and SimpleTech [re]drive attached. With her two USB drives, BiNG didn't freeze, recognized both drives, displayed their partitions and offered to let me resize, format, or delete them. Because they were not mine and had data on them, I didn't actually test these commands, but they were available and BiNG wasn't locked up as it was when the AcomData was attached and turned ON.

So, after loading the Windows desktop on ANY version of windows with 1.x or 2.0 USB ports, the AcomData works, flawlessly. But turn it ON anywhere else (outside of windows) and everything locks up until it is turned back OFF. On her system, turning it OFF enables one to continue (as if nothing happened), but on MY much older system things stay frozen until the "reset" button is pressed.

So BiNG has issues on both rigs. For complete clarity you are obviously running BiNG from a floppy ie to the dos interface.

I guess that only leaves the external itself as the culprit ie chipset issue or perhaps the partitions on the external are corrupted in some way (can't see how it would give the results you have though).

The way I would have proceeded if it were my external hard drive is as follows. 1. As per Multibooter, try another usb drive (disk or flash) on your rig and see whether the problem remains (and use BiNG also). 2. Remove the hard drive from the AcomData and replace with another and test.

Have a look at Puppy, it's not difficult and won't take up any time. PuppyLinux.com. It's a small download 95Mb (make sure that you check the MD5) burn to ISO and just boot from the CD (For version go for a retro version with the older kernel say v4.1). Connect you external to Puppy at the Desktop and you will see it appear on the desktop (sda1). Find GParted in the menu and it will show the partitions. Make sure you unmount any drive prior to removal or shutdown (right click unmount, mounted drives have a green spot on them). You will also be able to access all files and folders on ANY hard disk on the pc. Great for solving issues.

As a last thought on this. Your usb enclosure has a chipset (I have a Nexstar3 IDE with chipset Cypruss AL2PT) and perhaps your enclosure's chipset is the culprit and the common denominator.

I appreciate that this is an irritating problem for you. It would drive me nuts also.

Good Luck

PS. when connecting the external are any other usb peripherals connected.

This post has been edited by risk_reversal: 19 March 2009 - 03:41 AM

[joe tweaker]

Multibooter, on Mar 19 2009, 02:35 AM, said:

joe tweaker, on Mar 17 2009, 03:32 PM, said:

The problem is if I forget to STOP the external drive and POWER IT DOWN before every reboot, it is causing my system to crash with a "Fatal Exception 0E" at 0028: FF0328FC in VXD USBD(05) +0000169C when I reboot.

If I understand right, you have a shutdown problem when your Acomdata USB drive is connected. Have you tried ScanDisk on your USB drive?

It doesn't crash while it's shutting 98se down. The crash occurs on the restart, just before the windows desktop should appear, AND ONLY WHEN THE DRIVE REMAINS POWERED ON DURING THE ENTIRE RESTART. When the drive is OFF, windows restarts without any complaints.

The drive has NOT caused any abrupt or unexpected errors WHEN IT IS SWITCHED ON AFTER WINDOWS HAS LOADED. Nor has it ever lost any data, or caused windows to shut down while using it.

Fact is, at this point, I have not even copied any important data to it, other than for testing purposes. I copied a video to it and then played the video back from the USB drive as a test. And it played the video smoothly and without errors or stuttering.

All I desire, that I do not currently have, is to be able to leave the drive POWERED ON when I restart windows, without seeing a BSOD just before the desktop should reappear.

A quick scandisk says each partition has no bad blocks and is ready for data storage. I have not yet run a thorough surface scan due to the the time required and because it has not lost any test data copied to it. Since I haven't done this yet, I will start a surface scan on the first logical drive when I go to bed tonight, and let it run while I sleep. But I really don't expect to find any surface defects.

[risk_reversal]

Following your post I think that you can probably scrub out most of my earlier reply.

Interesting though that your external enclosure is problematic on your pc and your friend's when the drive is connected during the boot process (hope I understood that correctly). If so try another usb device as per Multibooter's suggestion and boot system to see if error repeats.

Good Luck

This post has been edited by risk_reversal: 19 March 2009 - 03:51 AM

[joe tweaker]

risk_reversal, on Mar 19 2009, 04:36 AM, said:

So BiNG has issues on both rigs. For complete clarity you are obviously running BiNG from a floppy ie to the dos interface.

I guess that only leaves the external itself as the culprit ie chipset issue or perhaps the partitions on the external are corrupted in some way (can't see how it would give the results you have though).

PS. when connecting the external are any other usb peripherals connected.

Yes, I am running BiNG from a floppy on my friend's computer. On mine it is installed on the hard disk, but I get the same result on my computer (AcomData not visible on USB 1.x ports) from the hard drive or the floppy.

I'm guessing it's a chipset issue at this point, but I'll probably never be 100% certain.

I have no other USB devices except for a 128MB Memorex Flash Thumb Drive, and it has never been plugged in during any tests on the AcomData. I don't think I've ever performed a restart with the Thumb Drive plugged in. I'll give that a test tonight, see what it does, before I start the surface scan.

[Multibooter]

joe tweaker, on Mar 19 2009, 01:58 AM, said:

I'll give that a test tonight, see what it does, before I start the surface scan.

I doubt the surface scan will be of much use. How about clearing the USB disk & then partitioning it into a single logical partition, 160GB is Ok on USB? My partitioning tool of choice for Win98/FAT32 is PartitionMagic 8.01, which can partition external USB drives.

My next step would be try the shutdown/restart sequence with another make of external USB HDD.

You could also check whether it's the fault of nusb, which I would doubt:
1) with your Acomdata USB drive connected, rename Windows\INF\USBSTOR.INF and USBSTOR.PNF to something like\Usbstor.inf.deactivated and usbstor.PNF.deactivated
New devices will not be detected by nusb anymore
2) go into Control Panel -> System -> Device Manager
select (probably under Disk Drives) the Acomdata USB drive -> click on Remove
3) reboot
4) install the Win98 driver from the Acomdata CD
5) reboot again, when your system is up again, plug in your Acomdata USB drive, Win98 should detect it Ok
6) shutdown/restart again, with the Acomdata drive connected: do you still have the same problem?

To re-activate nusb for installing other devices: just rename the renamed USBSTOR.INF & USBSTOR.PNF to their original names.

This post has been edited by Multibooter: 19 March 2009 - 09:17 PM

[dencorso]

Well, despite the error message, USBD is not a VxD, but a WDM driver. BTW, what version of USBD.SYS do you have in your system?

[joe tweaker]

dencorso, on Mar 19 2009, 10:37 PM, said:

Well, despite the error message, USBD is not a VxD, but a WDM driver. BTW, what version of USBD.SYS do you have in your system?

USBD.SYS 18,928 bytes 09/08/1999 12:46PM <-- version is 4.10.2224

98se restarts without incident when the Memorex Thumb Drive is left in a USB port during a restart.

Scandisk reports no surface defects in the first logical drive on the AcomData.

[jaclaz]

If I get it right, the PC tries, no matter what the settings in BIOS are to boot from the external USB if it is switched on at boot time.

If this is the case, it is possible to workaround the problem.

Which exact "crash" do you experience (are you getting a readable error - if yes post it EXACTLY as you see it - or a flashing cursor or j in the top left of screen or whatever - please describe it).

One can install on the external USB stick a MBR/Bootsector or loader that actually boots and then passes control to the internal drive.

Tests must be made to check whether drive numbers are shifted and a disk BIOS swap is needed or not.

This technique is used commonly for the "Install XP from USB" using grub4dos.

jaclaz

[joe tweaker]

Multibooter, on Mar 19 2009, 10:07 PM, said:

How about clearing the USB disk & then partitioning it into a single logical partition, 160GB is Ok on USB? My partitioning tool of choice for Win98/FAT32 is PartitionMagic 8.01, which can partition external USB drives.

But I don't have PM 8.01. I haven't used PM since v3.0. I use BiNG for most jobs now,
but for some reason this AcomData unit causes EVERY system I've tested to LOCKUP
when it's powered ON outside of windows. Within a windows environment, it's A-OK.

But outside windows (ie, DOS and BiNG) it's instant system LOCKUP. So it appears the
only hope of resizing it is with a partitioning tool that handles USB from within windows,
OR removing the disk from the AcomData enclosure and installing it on an IDE cable
in a system with a newer BIOS than mine (ie, one that can "see" a drive > 128GB).
Unfortunately, I don't have such a system available whenever I need one!

On my ABIT mb it seems the BIOS detects anything > 128GB as a 128GB disk.
Apparently the BM6 BIOS also limits what BiNG can do, because BiNG would not
permit more than 128GB to be used when it partitioned the 160GB disk.