VirtualBox

Opened 15 years ago

Closed 14 years ago

Last modified 14 years ago

#4627 closed defect (fixed)

Crazy keyboard repeating

Reported by: Sam Morris Owned by:
Component: other Version: VirtualBox 3.0.2
Keywords: Cc:
Guest type: Linux Host type: Windows

Description

After my VM has been running for some time, the keybaord goes a bit crazy. It begins to endlessly autorepeat whatever key was last pressed. e.g., I may press 'a', then KeyPress and KeyRelease events are generated for the 'a' key, forever.

The only way to fix this is to reboot the VM.

The host is Windows Vista Home Basic (64-bit) and guest is Debian using Linux 2.6.30.

I don't remember seeing this bug with versions prior to 3.0, BTW.

Attachments (1)

wintermute-2009-07-26-13-24-26.log (48.6 KB ) - added by Sam Morris 15 years ago.
virtualbox log

Download all attachments as: .zip

Change History (7)

by Sam Morris, 15 years ago

virtualbox log

comment:1 by Sam Morris, 15 years ago

xev reveals some details about what's going on.

When I *release* a keyboard key, the following event is generated:

KeyPress event, serial 33, synthetic NO, window 0x4a00001,
    root 0x109, subw 0x0, time 6038250, (86,-11), root:(203,514),
    state 0x0, keycode 32 (keysym 0x6f, o), same_screen YES,
    XLookupString gives 1 bytes: (6f) "o"
    XmbLookupString gives 1 bytes: (6f) "o"
    XFilterEvent returns: False

but the corresponding KeyRelease event does not occur until I *press* another key:

KeyRelease event, serial 33, synthetic NO, window 0x4a00001,
    root 0x109, subw 0x0, time 6104687, (34,-18), root:(151,507),
    state 0x0, keycode 36 (keysym 0xff0d, Return), same_screen YES,
"   XLookupString gives 1 bytes: (0d) "
    XFilterEvent returns: False

The KeyPress event from the second key only happens when I *release* it:

KeyPress event, serial 33, synthetic NO, window 0x4a00001,
    root 0x109, subw 0x0, time 6104901, (34,-18), root:(151,507),
    state 0x0, keycode 28 (keysym 0x74, t), same_screen YES,
    XLookupString gives 1 bytes: (74) "t"
    XmbLookupString gives 1 bytes: (74) "t"
    XFilterEvent returns: False

and so on.

Here are the corresponding events from /dev/input/event. Releasing a key:

Event: time 1249151610.266018, type 4 (Misc), code 4 (ScanCode), value 1f
Event: time 1249151610.266243, type 1 (Key), code 31 (S), value 1

Then pressing another key:

Event: time 1249151691.278281, type 4 (Misc), code 4 (ScanCode), value 1f
Event: time 1249151691.278412, type 1 (Key), code 31 (S), value 0

and releasing it:

Event: time 1249151691.509644, type 4 (Misc), code 4 (ScanCode), value 1e
Event: time 1249151691.509767, type 1 (Key), code 30 (A), value 1

comment:2 by Sam Morris, 15 years ago

BTW, I have a USB keyboard. I don't have any USB filters set up for this VM however.

I just noticed that I totally screwed up the xev output that I pasted in my previous comment. Anyway, I hope that what is going on is clearly demonstrated by the evtest output.

comment:3 by Sam Morris, 15 years ago

Still present in 3.0.4.

comment:4 by Sam Morris, 14 years ago

I've not seen this so far in 3.0.8.

comment:5 by Sander van Leeuwen, 14 years ago

Resolution: fixed
Status: newclosed

Thanks for the feedback. Note that 3.0.10 contains additional fixes for SMP guests.

comment:6 by Technologov, 14 years ago

This bug looks related to Bug #5408.

-Technologov

Note: See TracTickets for help on using tickets.

© 2023 Oracle
ContactPrivacy policyTerms of Use