[BUG] Occasional X_LINK_ERROR on POE devices #845
Labels
bug
Something isn't working
Comments
|
Run the WRT vibration - yes that might be the case if connection temporarily drops. Both issues can show themselves in similar light |
|
We're having trouble getting crash dumps out of the camera. We're running the latest release however often after our x_link_error there appears to be no crash dump available. Are the crash dumps stored in flash or ram? If so do they stick around after a power cycle and when do they get overwritten? |
|
The unhandled throw leading to terminate could be luxonis/XLink#59, the bugs I fix in #809, or the bug I'm waiting on Luxonis to answer #809 (comment) |
moratom
pushed a commit
that referenced
this issue
Apr 3, 2024
* Added stress test to cam_test.py * Default cam test: Stream from all connected cameras. If the device is calibrated and a stero pair is available, create a StereoDepth node and stream depth aswell * Added a --no-stereo option * improve stability of stress test, add dot projector and flood light brightness prints * Add LR support * Visualize outputs (not pretty) * Added flood light and dot projector to stress test * Only create one video encoder. Visualize tof * Limit to one edge detector. add exposure time and iso settings * fix clamp * zero edge detectors * no yolo * add another video encoder * Add all sockets * Add back yolo to stress test, added args to stress test * Forward stress test help to stress_test.py without being consumed by cam_test.py * Default no stereo, convert --no-stereo into --stereo option. Fix stress test index out of range when no resolution config is available. fix stress test stereo depth colorization * Added --yolo option to cam_test for ability to stress the device a bit more, while retaining full size streams * Modify stress test to not use stereo if calibration isn't available. * Lower frame pool sizes * Move setNumFramesPool * Pool sizes * Visualize disparity * Fix cam_test.py after merge with develop
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We are repeatedly getting occasional disconnects from our oakd-w-poe devices that reads with the following error. The disconnects seems to happen randomly or just after some time.
The error:
"terminate called after throwing an instance of 'std::runtime_error'
what(): Communication exception - possible device error/misconfiguration. Original message 'Couldn't read data from stream: 'depth' (X_LINK_ERROR)'
[oakd-11] process has died [pid 1499128, exit code -6, cmd /home/grover/catkin_ws/devel/lib/grover_vision/oakd_node __name:=oakd __log:=/home/grover/.ros/log/659fb568-d1e9-11ed-b878-23da073fdadf/oakd-11.log].
log file: /home/grover/.ros/log/659fb568-d1e9-11ed-b878-23da073fdadf/oakd-11*.log"
We noticed the same error on our usb devices occasionally, we put that down to vibration/usb-cable issues.
What could be causing this? Could vibration be causing Ethernet disconnect? It is using the Luxonis cable and we've confirmed the POE switch the camera is connected to isn't losing power. Could it be an issue with the pipeline?
The text was updated successfully, but these errors were encountered: