+++ This bug was initially created as a clone of Bug #8444 +++ Distribution: Debian Unstable This problem is reported on a few sites, such as http://www.houseofhelp.com/forums/showthread.php?t=61435&page=6 and https://bugs.launchpad.net/ubuntu/+source/linux/+bug/261710. This seems to happen with the 16GB and the 8GB stick. I have the 8GB. The syslog: Dec 10 07:22:58 dorian kernel: [ 3340.624010] usb 4-6: new high speed USB device using ehci_hcd and address 9 Dec 10 07:22:58 dorian kernel: [ 3341.056377] usb 4-6: configuration #1 chosen from 1 choice Dec 10 07:22:58 dorian kernel: [ 3341.056953] scsi6 : SCSI emulation for USB Mass Storage devices Dec 10 07:22:58 dorian kernel: [ 3341.057085] usb-storage: device found at 9 Dec 10 07:22:58 dorian kernel: [ 3341.057087] usb-storage: waiting for device to settle before scanning Dec 10 07:22:58 dorian kernel: [ 3341.057676] usb 4-6: New USB device found, idVendor=1b1c, idProduct=0ad0 Dec 10 07:22:58 dorian kernel: [ 3341.057679] usb 4-6: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Dec 10 07:22:58 dorian kernel: [ 3341.057680] usb 4-6: Product: Flash Voyager Dec 10 07:22:58 dorian kernel: [ 3341.057681] usb 4-6: Manufacturer: Corsair Dec 10 07:22:58 dorian kernel: [ 3341.057682] usb 4-6: SerialNumber: 00080221ECAED44C00CE Dec 10 07:23:03 dorian kernel: [ 3346.056152] usb-storage: device scan complete Dec 10 07:23:09 dorian kernel: [ 3351.668008] usb 4-6: reset high speed USB device using ehci_hcd and address 9 Dec 10 07:23:24 dorian kernel: [ 3366.780009] usb 4-6: device descriptor read/64, error -110 Dec 10 07:23:39 dorian kernel: [ 3381.996008] usb 4-6: device descriptor read/64, error -110 Dec 10 07:23:39 dorian kernel: [ 3382.212008] usb 4-6: reset high speed USB device using ehci_hcd and address 9 Dec 10 07:23:54 dorian kernel: [ 3397.324012] usb 4-6: device descriptor read/64, error -110 ...and so on. I am running a self compiled 2.6.27.8. There is a workaround given on the Launchpad site editinhg /etc/modprobe.d/options to increase the timeout which worked for me until I compiled the new kernel with the scsi modules built in and not as modules. Anyway it is annoying having to reinstall a kernel if I want to use the stick on other linux machines. By the way, the same happens with linux in a virtual machine. The kernel bug linked seems to be fixed in 2007. But the patch applied then is not applicable now. The code of hub_port_init has changed.
The values in /etc/modprobe.d/options apply only to driver built as modules. If scsi_mod is built into the kernel instead of as a module, you can get the equivalent effect by adding the option on the kernel's boot command line (in grub or whatever boot loader you use): scsi_mod.inq_timeout=20 What relation is there between the Corsair Flash Voyager and Bug #8444?
Bug #8444 is about the Corsair Flash Voyager, too. (And fixed by setting a timeout, and that the Flash Voyager works fine under Windows.) I just trusted the launchpad guys where I have the link to #8444 from and assumed that this issue is at least kind of similar. Even the logs are similar. Sorry if this is not the case.
As far as I can tell, these are two separate problems. That's judging from the fact that they have two separate solutions. Anyway, if the advice in comment #1 fixes your problem, you can close out this bug report.
Isn't there a way to fix it so that the Corsair Flash Voyager owners don't have to reboot the machine or rebuild the kernel of a machine where they want to use the memory stick? Would be a pity. And would be great if it was possible to fix it.
On Thu, 11 Dec 2008 bugme-daemon@bugzilla.kernel.org wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=12188 > ------- Comment #4 from paul@purecodes.org 2008-12-11 23:38 ------- > Isn't there a way to fix it so that the Corsair Flash Voyager owners don't > have > to reboot the machine or rebuild the kernel of a machine where they want to > use > the memory stick? > Would be a pity. And would be great if it was possible to fix it. Would anybody object if the default value for scsi_inq_timeout was increased? There is at least one USB device for which the current 5 seconds is known to be too small. Windows XP uses 20 seconds (for USB mass-storage devices, anyway -- I don't know about other bus types). Is there any reason not to make 20 the default? Alan Stern
All USB bugs should be sent to the linux-usb@vger.kernel.org mailing list, and not entered into bugzilla. Please bring this issue up there, if it is still a problem in the latest kernel release.