Bugzilla – Bug 140904
K3B doesn't find USB DVD writer
Last modified: 2007-01-19 10:31:35 UTC
K3B doesn't find the USB DVD writer since connected to USB 2.0 via add-on card insteadof built-in USB 1.0 card. This is true for both, normal users and root. K3B's configuration dialog doesn't offer the USB device. When starting as root, K3B emits: k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. k3b: ERROR: (K3bDevice::Device) Unable to do inquiry. The problem may be related to the syslog message: "kernel: usb 1-3: device not accepting address 33, error -71"
Is the DVD drive usable at all?
Not as a normal user, but as root (after "su", using k3b), and only after I unloaded the USB2 driver (that's another problem I know).
PLease attach the dmesg output: - after a reboot - after an access to the DVD as a user (without k3b) - after an access with k3b (running as root)
It seems the message "k3b: ERROR: (K3bDevice::Device) Unable to do inquiry." appearing four times is related to no media being in the DVD writer. When unloading "ehci_hcd" as root before the normal user logs in, k3b works for the normal user. However it only works with USB 1.0 speed then.
Created attachment 90098 [details] Output of "dmesg" (PC booted, ehci_hcd unloaded, k3b started)
did you upgrade this system to 10.1 or 10.2 alreeeady? If so, does it work better there? If it is still running 10.0, can you install the 10.1 or sles10 kernel on it to see if it makes any difference??
Some words: I have used the same DVD-Writer on a different PC running openSUSE 10.1 (kernel 2.6.16.27-0.6-default) with USB 1 (which is no fun of course). As it seems, at some point of idle time, the PC wants to make some power saving and logs into syslog: gnome-power-manager: Suspending computer because the system state is idle kernel: usbcore: deregistering driver usb-storage [...] kernel: uhci_hcd 0000:00:1f.2: USB bus 2 deregistered [powersave]: WARNING (executeScript:590) Program to trigger sleep (/usr/lib/powersave/do_acpi_sleep) failed and exited with status 127 kernel: USB Universal Host Controller Interface driver v2.3 Then everything (USB) reconnects, but the DVD Writer is missing in K3B then. That's probably a resmgr issue. I also added that USB 2.0 add-on card with unchanged results: The drive light comes on when connecting, but actually no connection could be established. Syslog says something like: kernel: usb 1-4: new high speed USB device using ehci_hcd and address 2 kernel: usb 1-4: device descriptor read/64, error -71 While trying the USB writer had locked up in a way that I could only connect it to USB1 after power cycling it.
I have good news: After replacing the USB cable with a shorter one, USB2 works! (As it seems the longest USB2 certified cables are 1.80m, while mine was 3m. Not that I actually needed that length, but that one was the only one on stock here) I still believe the kernel message could have been better than: kernel: usb 1-2: new high speed USB device using ehci_hcd and address 9 kernel: usb 1-2: device not accepting address 9, error -71
so, k3b can be used now? The usb driver has most likely no chance to know the length of the cable.
well, that kernel message shows pretty clear that it had troubles communicating with the device.
Regarding comment #9 and #10: Yes the device works fone now. About the cable issue: Windows/XP for example tells the user in cleartext (just for example) that an USB port had been disabled because the device was consuming too much power. I don't know about the exact effects of lengthy USB cables, but "error -71" isn't too helpful. Comparing it to SCSI's domain validation, a better reaction of the driver would be something like "high speed USB failed a test, trying full speed instead"... You get the idea. It's OK to close the bug however.
I have seen such a message in newer kernels, maybe it does not trigger for all cases.