wg-perception / object_recognition_ros

ROS components usable byt the object_recognition pipeline
16 stars 31 forks source link

Actionlib server crashed with linemod #21

Open xMutzelx opened 8 years ago

xMutzelx commented 8 years ago

Hello community, I get a strange error. I can't use the actionlib server/client with linemod. Tabletop works fine. First I start the server:rosrun object_recognition_ros server -crospack find object_recognition_linemod/conf/detection.ros.ork Second I start the client: rosrun object_recognition_ros client And this is my output:

[ INFO] [1477560425.096815042]: Connected to master at [localhost:11311]
[ INFO] [1477560425.100454190]: Initialized ROS. node_name: /object_recognition
[ INFO] [1477560425.545173251]: System already initialized. node_name: /object_recognition
[ INFO] [1477560425.810228513]: System already initialized. node_name: /object_recognition
[INFO] [WallTime: 1477560425.813119] ORK server configured
[INFO] [WallTime: 1477560425.817559] Subscribed to the recognized_object_array topic.
[INFO] [WallTime: 1477560425.831282] ORK server started
[ INFO] [1477560473.372871621]: Subscribed to topic:/camera/rgb/camera_info [queue_size: 1][tcp_nodelay: 0]
[ INFO] [1477560473.376344520]: Subscribed to topic:/camera/depth_registered/camera_info [queue_size: 1][tcp_nodelay: 0]
[ INFO] [1477560473.379594144]: Subscribed to topic:/camera/rgb/image_rect_color [queue_size: 1][tcp_nodelay: 0]
[ INFO] [1477560473.383015854]: Subscribed to topic:/camera/depth_registered/image_raw [queue_size: 1][tcp_nodelay: 0]
Reading the mesh file mesh.stl
Segmentation fault (core dumped)

After the detection failed, it creates an x�'��.stl (invalid encoding) file in my home archive. Any ideas what can cause this problem? Thank you very much for your help. If I use linemod continuous it works fine. Only the actionlib server/client result in this error. I am using Ubuntu 14.04 and ros indigo.

xMutzelx commented 8 years ago

GDB gives me this output:

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fff61d35700 (LWP 10995)]
0x00007fffe8965189 in glDeleteTextures () from /usr/lib/nvidia-304/libGL.so.1

I tried it again with my integrated IntelHD chip and it works. I updated my NVidia driver and it didn't help. Do you think this is an bug in ORK/Ecto or in the NVidia driver?

hris2003 commented 8 years ago

Or maybe it's all about the libGL?

On Fri, Oct 28, 2016 at 10:39 AM, xMutzelx notifications@github.com wrote:

GDB gives me this output:

Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fff61d35700 (LWP 10995)] 0x00007fffe8965189 in glDeleteTextures () from /usr/lib/nvidia-304/libGL.so.1

I tried it again with my integrated IntelHD chip and it works. I updated my NVidia driver and it didn't help. Do you think this is an bug in ORK/Ecto or in the NVidia driver?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/wg-perception/object_recognition_ros/issues/21#issuecomment-256867191, or mute the thread https://github.com/notifications/unsubscribe-auth/ACD8jhQU-xANVemxQo-dxYPG9qOWHQ5rks5q4bTTgaJpZM4KiLK3 .