I have been using a SpaceTraveller3 USB on a RHEL5u5 x86_64 system.
Today I replaced it with a SpaceNavigator and 3DxWare will no longer load - it segfaults! I have tried upgrading to the latest drivers (from 1.2.17 to 1.3) and plugging the old spaceball back in but it still segfaults.
I can run the driver manually but my application (based on the SDK) stops receiving updates from the spaceball after a short time.
I have managed to replicate this bug on another system.
This is an Intel dual-core system running CentOS 5.1 x86_64. It has had a SpaceTraveller3 USB running for over six months with out any configuration changes or software updates. I unplugged the SpaceTraveller yesterday and when I plugged it in again this morning it lit up but the 3DxWare window did not appear. /var/log/messages reveals that the driver began segfaulting when the SpaceTraveller was plugged back in. The driver runs manually but will no longer run from inittab!
The original machine has now had a fresh install of its operating system but 3DxWare will not run from inittab.
Please suggest how I might fix this problem. I am not sure where to go from here.
I have been using a SpaceTraveller3 USB on a RHEL5u5 x86_64 system.
Today I replaced it with a SpaceNavigator and 3DxWare will no longer load - it segfaults! I have tried upgrading to the latest drivers (from 1.2.17 to 1.3) and plugging the old spaceball back in but it still segfaults.
I can run the driver manually as root but it wont run from the inittab.
uludwig wrote:do you have a log file from the 3DxWareUnix? (/tmp/3DxWare.log) It is created if you add the options -debug -E to the entry in the inittab.
Do you have more information there? Do you get a core file?
Both systems produce the same log output which you can see in my original post. The device lights up and then the driver segfaults.
I don't think it created a corefile, where should I be looking? How can I make it produce one when it is being run from inittab?
uludwig wrote:From what I see from the log file it looks as if the authorization at the XServer causes problems. Do you see any difference if you do a
xhost +
from the command line?
Uta
I tried "xhost +" this morning on both systems but the driver still segfaults. I have installed CentOS 5.2 x86_64 on one of these systems now and it has the same problem with a fresh install!
Aug 16 07:41:55 test kernel: input: USB HID v1.10 Multi-Axis Controller [3Dconnexion SpaceTraveler USB] on usb-0000:00:1d.0-1
Aug 16 07:42:06 test kernel: usb 5-1: usbfs: process 16430 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:06 test kernel: 3dxsrv[16430]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff2b14ceb0 error 6
Aug 16 07:42:07 test kernel: usb 5-1: usbfs: process 16437 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:07 test kernel: 3dxsrv[16437]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff6f146eb0 error 6
Aug 16 07:42:08 test kernel: usb 5-1: usbfs: process 16458 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:08 test kernel: 3dxsrv[16458]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff4715dec0 error 6
Aug 16 07:42:09 test kernel: usb 5-1: usbfs: process 16468 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:09 test kernel: 3dxsrv[16468]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff1dc03960 error 6
Aug 16 07:42:10 test kernel: usb 5-1: usbfs: process 16478 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:10 test kernel: 3dxsrv[16478]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff0f3450b0 error 6
Aug 16 07:42:11 test kernel: usb 5-1: usbfs: process 16488 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:11 test kernel: 3dxsrv[16488]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff12bfd960 error 6
Aug 16 07:42:12 test kernel: usb 5-1: usbfs: process 16498 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:12 test kernel: 3dxsrv[16498]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff6a41f180 error 6
Aug 16 07:42:13 test kernel: usb 5-1: usbfs: process 16508 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:13 test kernel: 3dxsrv[16508]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fff113460b0 error 6
Aug 16 07:42:14 test kernel: usb 5-1: usbfs: process 16518 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:14 test kernel: 3dxsrv[16518]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fffbce1fb80 error 6
Aug 16 07:42:15 test kernel: usb 5-1: usbfs: process 16527 (3dxsrv) did not claim interface 0 before use
Aug 16 07:42:15 test kernel: 3dxsrv[16527]: segfault at 0000000000000010 rip 0000003bb3471966 rsp 00007fffa3d99b00 error 6
Aug 16 07:42:15 test init: Id "3d" respawning too fast: disabled for 5 minutes
uludwig wrote:do you see any difference when you create a new user?
These computers use a network login so it is possible that the process launched from inittab is unable to access the X authority file. I will try and test this.
As of this morning, the driver is loading on my workstation again. I am not sure why so I will have to try and get it to fail again.