"Failed to create CaptureSession" when continuously capturing frames from cameras

[xzry66]

Hi,

We have a project using Jetson Xavier AGX and MIPI cameras. We’re using Libargus to connect to MIPI cameras. Currently we’re testing camera stability by continuously capturing frames from each camera. We’ve observed a few times that right after 1014 loops, libargus begin to output following error and cannot connect to cameras anymore unless restart AGX.

(Argus) Error Timeout:  (propagating from src/rpc/socket/client/ClientSocketManager.cpp, function send(), line 137)
(Argus) Error Timeout:  (propagating from src/rpc/socket/client/SocketClientDispatch.cpp, function dispatch(), line 91)
Error generated. main.cpp, initialize:129 Failed to create CaptureSession
Error generated. main.cpp, execute:505 Failed to initialize Camera session 0
(Argus) Error Timeout:  (propagating from src/rpc/socket/client/ClientSocketManager.cpp, function send(), line 137)
(Argus) Error Timeout:  (propagating from src/rpc/socket/client/SocketClientDispatch.cpp, function dispatch(), line 91)
(Argus) Error InvalidState: Argus client is exiting with 2 outstanding client threads (in src/rpc/socket/client/ClientSocketManager.cpp, function recvThreadCore(), line 357)

We’re testing on Jetpack 4.6.2 [L4T 32.7.2] on Jetson Xavier AGX.
The method we capture camera frame referring to jetson_multimedia_api/samples, basically we start the camera, then capture one frame and stop the camera and run service nvargus-daemon restart.

Not sure why this happens exactly after 1014 loops. Wonder if resource has been fully cleaned up after stopping the camera and running service nvargus-daemon restart.

[ShaneCCC]

Can you share the script to reproduce the issue.

Also please run the nvargus-daemon manually to check the log if more clue.

sudo service nvargus-daemon stop
sudo su
nvargus-daemon

run the sample in another terminal