I'm currently on a fully up2date RHL 8.0 system w/ 2 PII CPUs running the kernel-smp-2.4.18-19.8.0 kernel. Sometimes, after periods of inactivity, my USB keyboard will be "hung". The LEDs will blink, and the keyboard will not respond to any key presses. Unplugging and replugging the keyboard in resets it and makes it responsive again. Messages displayed while the keyboard is hung: Jan 12 12:15:20 verdande kernel: usb-uhci.c: interrupt, status 3, frame# 920 Jan 12 12:15:25 verdande kernel: usb-uhci.c: interrupt, status 3, frame# 1864 Jan 12 12:15:30 verdande kernel: usb-uhci.c: interrupt, status 3, frame# 760 Jan 12 12:15:35 verdande kernel: usb-uhci.c: interrupt, status 3, frame# 1704 every 5 seconds then unplug and replug: Jan 12 12:15:39 verdande kernel: usb.c: USB disconnect on device 3 Jan 12 12:15:40 verdande kernel: hub.c: USB new device connect on bus1/2, assign ed device number 4 Jan 12 12:15:40 verdande kernel: input0: USB HID v1.00 Keyboard [QTRONIX USB Key board and Mouse] on usb1:4.0 Jan 12 12:15:40 verdande kernel: input1: USB HID v1.00 Mouse [QTRONIX USB Keyboa rd and Mouse] on usb1:4.1 Jan 12 12:15:44 verdande /etc/hotplug/usb.agent: Setup hid for USB product 5c7/2 011/110 Jan 12 12:15:44 verdande /etc/hotplug/usb.agent: Setup keybdev mousedev for USB product 5c7/2011/110 and it works again, at least until it hangs again. This isn't specific to this 19.8 kernel -- I saw it with older 8.0 kernels as well. It only seems to occur when the keyboard is idle for a while and only while in X (but I tend to always be in X on that system, so that may or may not be required; similarly, I presume its a kernel problem, but it might be an X problem since I've only seen it there) usbview describes the keyboard as: USB Keyboard and Mouse Manufacturer: QTRONIX Speed: 1.5Mb/s (low) USB Version: 1.00 Device Class: 00(>ifc ) Device Subclass: 00 Device Protocol: 00 Maximum Default Endpoint Size: 8 Number of Configurations: 1 Vendor Id: 05c7 Product Id: 2011 Revision Number: 1.10 Config Number: 1 Number of Interfaces: 2 Attributes: a0 MaxPower Needed: 50mA Interface Number: 0 Name: hid Alternate Number: 0 Class: 03(HID ) Sub Class: 1 Protocol: 1 Number of Endpoints: 1 Endpoint Address: 81 Direction: in Attribute: 3 Type: Int. Max Packet Size: 8 Interval: 20ms Interface Number: 1 Name: hid Alternate Number: 0 Class: 03(HID ) Sub Class: 1 Protocol: 2 Number of Endpoints: 1 Endpoint Address: 82 Direction: in Attribute: 3 Type: Int. Max Packet Size: 3 Interval: 10ms I'm using the usb-uhci controller module. This system's USB controller is: 00:07.2 USB Controller: Intel Corp. 82371AB/EB/MB PIIX4 USB (rev 01)
BTW, if it matters, this keyboard has an integrated PS/2 port, but I'm not using it. The mouse on this system is a serial mouse
Seeing this with kernel-2.4.20-8smp on RHL 9 as well
Does Fedora do the same?
Dax, what are you doing on cc:? Do you have a reproducer for this?
No, I was just curious.
alphas & betas up to the release of Fedora Core 1 did (Aug / Sep / Oct kernel releases) I've not tried any of the most recent Fedora kernels since I moved cross-country in November and that machine hasn't (yet). It should be getting here at the end of this month, though, so I'll update then....
Please retry with either FC1 or FC2 Test releases (2.6 kernel). Normally these symptoms point to BIOS trying to emulate legacy keyboard under our feet though, so I don't think I can do anything. But let's not discount a possible SMP breakage.
Still does it with FC1 (2179 kernel)
Have you tried a different keyboard? Trying to split hid off uhci here.
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/