USB drive mount disappears randomly, cannot remount (Red Hat 9)
 

(I am reposting this as my post in hardware got no responses after a week - so am hoping for some help here.)

I have a USB "bracket" around a plain old IDE hdd, it sits outside the pc. When I boot RedHat it detects it fine and it mounts fine.

I listen to mp3 tracks off that drive and it can work for hours or days when suddenly it stops responding and I cannot even do an ls on the drive.

I have tried unplugging the drive and then re-plugging it and doing a mount -a but I inevitably get a "mount: /dev/sda1 is not a valid block device" error.

I have tried rmmod usb-storage and then modprobe usb-storage but I am a newbie and clueless as to what those commands do - anyway they did squat.

If I do a tail /var/log/messages I get stuff like:
Dec 20 15:11:41 localhost kernel: usb.c: USB disconnect on device 00:10.3-3 address 4
Dec 20 15:11:41 localhost devlabel: devlabel service started/restarted
Dec 20 15:12:15 localhost kernel: hub.c: new USB device 00:10.3-3, assigned address 5
Dec 20 15:12:18 localhost /etc/hotplug/usb.agent: Setup usb-storage for USB product 7c4/ad01/104
Dec 20 15:12:19 localhost devlabel: devlabel service started/restarted

Where the USB device seems to get assigned address X, where X just keeps counting up. No amount of trying to mount against sdb, sdc, sdd seems to work - so I cannot fathom the connection between this "address" and the /dev/sd? nodes.

I would like the USB to stop failing, but if I could just have a little "recipe" to get it working again without restarting the machine that would be fine.

Specs:
fstab contains:
/dev/sda1 /mnt/usbdrv ntfs auto,ro,uid=donn 0 0
Drive is NTFS.
O/S RedHat 9.0
Kernel: 2.4.20-20.9
PC: AMD Athlon

Thanks.

have you update your kernel recently or did you do anything before it stops working?

i assume you have a device or your usb device which was named "Tux"
what you need to do to test it wether linux had detect it was to type "dmesg | grep Tux" in your terminal to found out about it.
also, you might try "cdrecord -scanbus" to see if your device was listed.

Tricky,
I have upgraded the RedHat kernel, but the USB drive did the same in the previous kernel.

I do nothing more than play music from the drive - it literally dissapears randomly, i.e. at any given moment - no help needed from me!

I did your cdrecord trick (wierd :)) pasted:
" 0,0,0 0) 'USB 2.0 ' 'Storage Device ' '0100' Disk "

What does this line mean ?

I am not sure what the device is called - so I grepped for "USB" :
dmesg | grep USB
ehci-hcd 00:10.3: VIA Technologies, Inc. USB 2.0
usb.c: new USB bus registered, assigned bus number 1
ehci-hcd 00:10.3: USB 2.0 enabled, EHCI 1.00, driver 2003-Jan-22
hub.c: USB hub found
usb-uhci.c: USB UHCI at I/O 0xd000, IRQ 10
usb.c: new USB bus registered, assigned bus number 2
hub.c: USB hub found
usb-uhci.c: USB UHCI at I/O 0xd400, IRQ 10
usb.c: new USB bus registered, assigned bus number 3
hub.c: USB hub found
usb-uhci.c: USB UHCI at I/O 0xd800, IRQ 11
usb.c: new USB bus registered, assigned bus number 4
hub.c: USB hub found
usb-uhci.c: v1.275:USB Universal Host Controller Interface driver
hid-core.c: USB HID support drivers
hub.c: new USB device 00:10.3-3, assigned address 2
usb.c: USB device 2 (vend/prod 0x7c4/0xad01) is not claimed by any active driver.
hub.c: new USB device 00:10.1-2, assigned address 2
usb.c: USB device 2 (vend/prod 0x4a5/0x20b0) is not claimed by any active driver.
scanner.c: USB scanner device (0x04a5/0x20b0) now attached to scanner0
scanner.c: 0.4.12:USB Scanner Driver
Initializing USB Mass Storage driver...
scsi0 : SCSI emulation for USB Mass Storage devices
Vendor: USB 2.0 Model: Storage Device Rev: 0100
WARNING: USB Mass Storage data integrity not assured
USB Mass Storage device found at 2
USB Mass Storage support registered.

Does this tell you something I am missing?
(Here's hoping)
:rolleyes:

scsi0 : SCSI emulation for USB Mass Storage devices
Vendor: USB 2.0 Model: Storage Device Rev: 0100
which tell you that if had recognize your drive and it's scsi emulated which was neccessary for your drive to operate.

just login as root don't do "su" but "su -"
$password---

$mkdir drive
$mount -t ntfs /dev/sda0 drive

see what it says.

I have tried that also, the error I get is always:
"mount: /dev/sda1 is not a valid block device"

I have also tried sda2 and sdb and sdb1 - all just guesses and none work.
If I switch the drive off and then on - Linux detects it again but I cannot seem to mount it anymore.
How do I know what "sd" to use when sda1 stops working?

did you check does it work in window?
the hardware might be fault! or broken

I have used the drive under Windows - but I have not left it on for hours and hours.
This is because I do not use Windows, so I don't have the time to test the drive under Windows for so long.

You may be right about this - I hope to have some time overnight maybe to test this.

I was hoping to find if others had solved this problem, so I could do the same.

Do you know the relationship between the scsi number (...assigned address 5 ...assigned address 6 ..etc) and the "sdx" directory names?

Thanks for your help though.

it looks like they are related to each other somehow. however, that's not the case. the are not related, but they are.
if see to drives listed using "cdrecord -scanbus"
then there is to optical drive connection to your computer. but both drive could be 1) /dev/hdc 2) /dev/hdd which wasn't scsi emulated.
if you did scsi emulated by add options to you boot loader then they should be
1)/dev/scd0 2)/dev/scd1

Tricky - I am not sure I understood your last post.
I must, unfortunately, go offline (I can only connect at weekends and nights) so if you have any other information - please post it and I will be back as soon as I can.

Thanks again.

do a "lsmod" after booting your linux box and please post the output.
so far, did you check is it working in window?
You might just wear out the drive or broke it by leaving it on so long.
Just want you to check the drive is still good.

Tricky-
I have not had time to test drive in Windows, but think about this: I leave the drive switched on; when it stops working, I reboot and Linux mounts it again fine.
I keep doing this, reboot=works, reboot=works, reboot=works, with no actual powering off the drive, so this indicates that the drive works 100% and the problem is Linux.

Here is my lsmod:
Module Size Used by Tainted: P
ppp_deflate 4440 0 (autoclean)
zlib_deflate 20536 0 (autoclean) [ppp_deflate]
ppp_async 9312 1 (autoclean)
ppp_generic 24188 3 (autoclean) [ppp_deflate ppp_async]
slhc 6660 0 (autoclean) [ppp_generic]
udf 97152 0 (autoclean)
via82cxxx_audio 24248 3 (autoclean)
uart401 8356 0 (autoclean) [via82cxxx_audio]
ac97_codec 14536 0 (autoclean) [via82cxxx_audio]
sound 73492 0 (autoclean) [via82cxxx_audio uart401]
soundcore 6468 4 (autoclean) [via82cxxx_audio sound]
agpgart 47968 3 (autoclean)
nvidia 1764992 11 (autoclean)
parport_pc 18788 1 (autoclean)
lp 8804 0 (autoclean)
parport 36800 1 (autoclean) [parport_pc lp]
autofs 13108 0 (autoclean) (unused)
ipt_REJECT 3992 6 (autoclean)
iptable_filter 2412 1 (autoclean)
ip_tables 14648 2 [ipt_REJECT iptable_filter]
sg 35852 0 (autoclean)
sr_mod 17720 2 (autoclean)
ide-scsi 11984 1
ide-cd 35424 1
cdrom 33216 0 [sr_mod ide-cd]
sd_mod 13484 2 (autoclean)
ntfs 58432 6 (autoclean)
nls_iso8859-1 3548 9 (autoclean)
nls_cp437 5180 1 (autoclean)
vfat 12780 1 (autoclean)
fat 38488 0 (autoclean) [vfat]
scanner 11736 0
usb-storage 73536 1
scsi_mod 106520 5 [sg sr_mod ide-scsi sd_mod usb-storage]
keybdev 2912 0 (unused)
mousedev 5428 1
hid 21892 0 (unused)
input 5792 0 [keybdev mousedev hid]
usb-uhci 25996 0 (unused)
ehci-hcd 19784 0 (unused)
usbcore 78272 1 [scanner usb-storage hid usb-uhci ehci-hcd]
ext3 69984 2
jbd 51508 2 [ext3]

Thankyou for your patience - I cannot be online for long during week days.

no-one out there.... :(