Issue with RTSP Streaming on Jetson Orin NX (Jetpack 5.1.2)

[user34150]

I am currently using a Jetson Orin NX with Jetpack 5.1.2 and facing an issue while using v4l2loopback to create a virtual video device (video3), which is then used for RTSP streaming.

Issue Description:

When an RTSP connection is first established to my device, it works fine. However, if I attempt to reconnect, my application crashes with the following logs:

stream ready at rtsp://127.0.0.1:8554/payload
RTSP[run] (): rtsp running 
Opening in BLOCKING MODE 
NvMMLiteOpen : Block : BlockType = 4 
===== NvVideo: NVENC =====
NvMMLiteBlockCreate : Block : BlockType = 4 
H264: Profile = 66, Level = 0 
NVMEDIA: Need to set EMC bandwidth : 846000 
Opening in BLOCKING MODE 
Failed to open /dev/video1: No such file or directory
Opening in BLOCKING MODE
/usr/bin/payload/app-rtsp/run_rtsp_app.sh: line 47: 14017 Segmentation fault      sudo ./rtsp_app

The issue is resolved when I stop the nvargus-daemon service using:

sudo systemctl stop nvargus-daemon.service

However, stopping this service causes image corruption in certain cases. If I do not stop nvargus-daemon, I receive the following logs:

minhvo@ubuntu:~$ sudo /usr/sbin/nvargus-daemon
=== NVIDIA Libargus Camera Service (0.99.33)=== Listening for connections...=== rtsp_app[14018]: Connection established (FFFF814BC900)OFParserListModules: module list: /proc/device-tree/tegra-camera-platform/modules/module0
OFParserListModules: module list: /proc/device-tree/tegra-camera-platform/modules/module1
NvPclHwGetModuleList: WARNING: Could not map module to ISP config string
NvPclHwGetModuleList: No module data found
NvPclHwGetModuleList: WARNING: Could not map module to ISP config string
NvPclHwGetModuleList: No module data found
OFParserGetVirtualDevice: NVIDIA Camera virtual enumerator not found in proc device-tree
---- imager: No override file found. ----
---- imager: No override file found. ----
(NvCamV4l2) Error ModuleNotPresent: V4L2Device not available (in /dvs/git/dirty/git-master_linux/camera/utils/nvcamv4l2/v4l2_device.cpp, function findDevice(), line 256)
(NvCamV4l2) Error ModuleNotPresent:  (propagating from /dvs/git/dirty/git-master_linux/camera/utils/nvcamv4l2/v4l2_device.cpp, function initialize(), line 60)
(NvOdmDevice) Error ModuleNotPresent:  (propagating from dvs/git/dirty/git-master_linux/camera-partner/imager/src/devices/V4L2SensorViCsi.cpp, function initialize(), line 111)
NvPclDriverInitializeData: Unable to initialize driver v4l2_sensor
NvPclInitializeDrivers: error: Failed to init camera sub module v4l2_sensor
NvPclStartPlatformDrivers: Failed to start module drivers
NvPclStateControllerOpen: Failed ImagerGUID 0. (error 0xA000E)
NvPclOpen: PCL Open Failed. Error: 0xf
SCF: Error BadParameter: Sensor could not be opened. (in src/services/capture/CaptureServiceDeviceSensor.cpp, function getSourceFromGuid(), line 726)
SCF: Error BadParameter:  (propagating from src/services/capture/CaptureService.cpp, function addSourceByGuid(), line 453)

Request for Assistance:

1. How can I resolve this issue without stopping nvargus-daemon?
2. If stopping nvargus-daemon is the only solution, what are the potential side effects on my system?
3. Is there an alternative approach to avoid segmentation faults while keeping nvargus-daemon running?

[user34150]

Previously, when using JetPack 4.6, no issues occurred. However, after upgrading to JetPack 5.1.2, the issue appeared.

[DaneLLL]

Hi,
If you don’t use Argus stack, please disable the daemon and remove the lib:

/usr/lib/aarch64-linux-gnu/tegra/libv4l2_nvargus.so

And check if the issue is still present.

[user34150]

Can you explain to me what is wrong with libv4l2_nvargus.so and if i disable it what parts i cant use?

[user34150]

The RTSP app issue has been resolved, but the image captured from the camera appears to have a problem. You can see the image below. It seems that disabling the library you mentioned is causing this error. Do you have any suggestions for a solution? I am capturing the camera using nvv4l2camerasrc.
Additionally, here are some dmesg errors I’d like to share with you, which I hope might be useful

[T3 Feb 11 10:37:39 2025] falcon 154c0000.nvenc: Direct firmware load for nvhost_nvenc080.fw failed with error -2
[T3 Feb 11 10:37:39 2025] falcon 154c0000.nvenc: Falling back to sysfs fallback for: nvhost_nvenc080.fw
[T3 Feb 11 10:37:39 2025] falcon 154c0000.nvenc: looking for firmware in subdirectory
[T3 Feb 11 10:37:42 2025] cpufreq: cpu4,cur:2102000,set:1984000,set ndiv:155
[T3 Feb 11 10:37:42 2025] pwm-tegra-tachometer 39c0000.tachometer: Tachometer Overflow is detected
[T3 Feb 11 10:37:45 2025] nvmap_alloc_handle: PID 2192: Payload-App: WARNING: All NvMap Allocations must have a tag to identify the subsystem allocating memory. Please pass the tag to the API call NvRmMemHanldeAllocAttr() or relevant.
[T3 Feb 11 10:37:45 2025] bwmgr API not supported
[T3 Feb 11 10:38:13 2025] nvgpu: 17000000.ga10b             railgate_enable_store:323  [INFO]  railgate is disabled.
[T3 Feb 11 10:39:54 2025] NVRM rpcRmApiControl_dce: NVRM_RPC_DCE: Failed RM ctrl call cmd:0x731341 result 0xffff:
[T3 Feb 11 10:39:57 2025] NVRM rpcRmApiControl_dce: NVRM_RPC_DCE: Failed RM ctrl call cmd:0x731341 result 0xffff:

[DaneLLL]

Hi,
Please share information about the camera. Not sure but probably it uses Argus stack.

[user34150]

We use a Sony camera block and custom driver, is there anything we need to note?

[DaneLLL]

Hi,
Please share the gstreamer command for launching the camera for reference. Would like to confirm which software stack it goes through.

[user34150]

this is my main pipeline then the rtsp app will read the video3

[DaneLLL]

Hi,
After removing libv4l2_nvargus.so, please clean gstreamer cache:

$ rm .cache/gstreamer-1.0/registry.aarch64.bin

And re-launch the pipelines. See if it works.

If it still does not work, please try v4l2src for video3. Would like to clarify if it is specific to nvv4l2camerasrc.

[user34150]

The issue has been resolved. The blue image was caused by the H265 decoder on the client machine; simply switching the server to H264 fixed it. Thank you for your support!