Bug 74635
Summary: | Sticky keyboard keys (get stuck on autorepeat) | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | berto <aguilr> |
Component: | kernel | Assignee: | Dave Jones <davej> |
Status: | CLOSED WONTFIX | QA Contact: | Brian Brock <bbrock> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 8.0 | CC: | bostjan, julo42, len.brown, mharris, pfrields, redhatbugzillaold, sven-redhat |
Target Milestone: | --- | Keywords: | Triaged |
Target Release: | --- | ||
Hardware: | i686 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2004-05-27 14:08:33 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 79579, 100644 |
Description
berto
2002-09-28 01:51:27 UTC
When you say the keyboard has stopped working, which of the following work or don't work: - using the window manager shortcuts (such as your Alt+1, or Alt+Tab) - typing in a terminal window - typing in other applications - Ctrl+Alt+F1 to switch to a virtual console The keyboard stops working as in typing in any application that takes normal keyboard input. I've tried typing in gnome-terminal, gaim, emacs ( GUI version ), evolution, xchat, and Galeon, but none of them work. Another interesting tidbit I forgot to mention is that when the keyboard does continue to work, if there is an active window ready to take input in that viewport, it will enter the value of the unmodified key. For example, the shortcut to my 1st viewport is winkey+1, if there is an active teminal window in the viewport when i switch to it, the character '1' will be typed in the window. Hope this helps a bit more. Forgot to mention that other window manager shortcuts such as alt+tab and switching to a virtual console with ctrl+alt+F1 still work fine. I have also made sure that it's not my keyboard; I checked that the connection to the back of the computer is secure and that my keys are working properly. They all seem to work without a problem. I would try a different keyboard, but I don't have an extra one around ... : \ Hmm, another strange thing happened just now too. I was holding down backspace to get rid of a word and when I let go of the key, it kept backspacing for another little bit before letting up. Also in Galeon, when i hit ctrl+pageup/pagedown to switch between tabs, it kept switching tabs spurratically, and when i hit ctrl+w to close a tab it sometimes closes more than one. It seems like the keyboard commands are 'sticking' somehow ... ? Can you open Preferences->Accessibility and be sure all that stuff is unchecked? I opened up the Accessibility dialog and the only thing that was checked was "Beep when enabling/disabling keyboard accessibility features". I have unchecked that, and the problem continues. If there are any config files you want me to attach, or a list of rpms installed, let me know. Thanks for all the help so far! Hmm. I don't have a lot more ideas offhand; would be interesting to see if upgrading to final released packages makes a difference. I upgraded to the final release of 8.0 and the keyboard still occasionally gets 'stuck'. I set the shortcuts to alt+1 - 6 for the 6 view ports. I'm not sure what other details are relevant. Just a quick note; I bought a new keyboard this weekend but am still getting 'sticky' keys. I was hoping that would have solved it. : \ I'm not sure how related they are, or if that was the problem, but i compiled and installed kernel 2.4.19 and my keyboard problems have gone away. I'm very happy. : ) Thanks, Roberto. I believe there's an upstream bug about this that may be enlightening, on http://bugzilla.gnome.org. Need to find said bug. I've seen this problem occasionally also, but it's hard to reproduce. Not sure we have much chance of making progress on this bug, but keeping it open for now. I am also seeing this bug. Seems like a race condition. My guess is that there are some tricky timing issues at the driver level and that 2.8GHz is just a bit too fast. For me this only happens on my Apple USB keyboard, not on my Sun USB keyboard. What kinds of keyboard do people have? Probably same as https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=74760 Yes this sounds like a dupe. "get stuck on autorepeat" is a new issue not mentioned in the other bug, however. I seem to have that problem, too... mostly when deleting Mails in evolution, it sometimes deletes more than one when pressing CTRL-D. Could this be related to the "software" keyrepeat that is active with RH8? I mean it looks like under X, the builtin keyboard controller keyrepeat is no longer used, otherwise you couldn't set a minimal delay of smaller than 250ms and repeats of more than 30keys/sec (great new feature btw.. I've always missed it!) *** Bug 74760 has been marked as a duplicate of this bug. *** Arjan - several people saying kernel upgrade fixed this on bug 74760 Havoc: I use a Happy Hacking Keyboard Lite (PS/2). And upgrading to the latest official RH8 kernel (kernel-2.4.18-19.8.0) didn't fix it. I don't want to upgrade to a rawhide 2.4.20 kernel on this production machine.. I believe the "get stuck on autorepeat" problem is the same as bug 66520. This happens to me, but it seems to only happen with modified or special keys: ctrl-t or ctrl-w will open or close between 1 and 3 tabs in galeon, for example, or ctrl-alt-arrow will move me one to three workspaces, or Ctrl-C will kill one or two applications in different places (quitting emacs, for example, also ends up killing whatever is running in the terminal that gets focus next). But I haven't seen it with regular letters. Update: I see this bug with a variety of keystrokes, mostly in combination, and mostly when the system is under moderate to high load. I am running a 500MHz PIII, so it's not a "my system is too fast" problem. I have just upgraded to kernel-2.4.18-27.8.0 and the issue is still there. This seems to have gone away after upgrading to a 2.4.20 kernel. I just wanted to tell you that I used to have this problem on my Fujitsu/Siemens Amilo D laptop using stock Red Hat 9. Sometimes, keys got stuck and I had to hit this key again to fix the problem. It used to happen with any key by the way (not only modifiers). And I noticed it mostly when using Evolution. Now, I installed Fedora Core Test 3 and the problem seems fixed. PS: I had filed this bug in bugzilla.redhat.com but I can't find it back :( I have sad news... Using fedora Core Test 3, I can reproduce this bug. But only if my laptop is un-plugged. The thing is that, if my laptop is unplugged, it uses its battery and runs more slowly. It seems then that, if the laptop runs more slowly, this bug appears. Now, I have also noticed something: When I use my laptop plugged, I have a strange bug: when using a lot of CPU (for example, running a 3D-accelerated game), my computer simply shuts down by itself... I didn't have this bug in Redhat 9. But I had the sticky keys bug. Now, if I un-plug my laptop, I don't have the auto-shutdown problem but I have the sticky keys one. Do you think those 2 bugs might be related ? PS: my laptop is a Fujitsu/Siemens Amilo D with an ATI Radeon M6. I have changed my laptop's BIOS setting to "smart throttle" power management. I don't know what it's supposed to be, but now my laptop is slower, even when plugged and the sticky keys bug is clearly back. See https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=107937 for more info. Another note: disabling ACPI (using acpi=off) makes the bug go away again. *** Bug 89702 has been marked as a duplicate of this bug. *** OK, after some investigation, I found out that the problem occurs when: - You have ACPI activated in your kernel (using acpi=on) - Whether you use acpid or not - The temperature of your CPU is too high In my case, a "cat /proc/acpi/thermal_zone/THR1/trip_points" gives the following: critical (S5): 78 C And the bug appears whenever the temperature reaches 79 C (according to "cat /proc/acpi/thermal_zone/THR1/temperature") I have tried to change the critical trip point to a higher temperature but the problem still occurs at 79 C... Does anyone have any idea about that ? Please open a new bug for the ACPI problem. Done. It's bug #110332. If you are experiencing the problem reported in this bug report, please read bug #76959 as it describes a similar problem. I have just added a patch to XFree86 which may fix that bug report, and it's possible it may fix this one also. If you find that it does fix the problem, please reassign this bug to XFree86, and close as a duplicate of bug #76959. Do the same with any other similar bug reports you're aware of, if you can confirm them as well. TIA |