Using multiple nvv4l2h264enc cause error

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) RTX 3090
• DeepStream Version 7.0
• JetPack Version (valid for Jetson only)
• TensorRT Version deepstream 7.0 installation guide
• NVIDIA GPU Driver Version (valid for GPU only) 535
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
hi i tried to save frames through multifile sink without any example exist in deepstream python binding but i got this error :
ENC_CTX(0x740984005590) Error in initializing nvenc context Error: gst-core-error-quark: ENCODER INITIALIZATION FAILED (1): None ENC_CTX(0x7408cc004f00) Error in initializing nvenc context ENC_CTX(0x7408e8004f10) Error in initializing nvenc context ENC_CTX(0x7408e0004f10) Error in initializing nvenc context [NvMultiObjectTracker] De-initialized

Your graph picture is too small to see the details of the pipeline. I guess you have more than 8 nvv4l2h264enc plugins in your pipeline. Please refer to Video Encode and Decode Support Matrix | NVIDIA Developer, the RTX 3090 only support less or equal to 8 encoder sessions.

then how can i add 16 camera using urisrcbin in my pipeline, is urisrcbin not using nvv4l2h264dec? because from the url you sent said the maximum number of decoder or encoder must be equal or less than 8

There is no limitation for decoder, only limitation for encoder. Please check Video Encode and Decode Support Matrix

then there is a way to use tee in urisrcbin before decoding ? similar to smart record pipeline

Are you talking about nvurisrcbin?

yes nvurisrcbin

nvurisrcbin is single pad output element. You can write your own source with basic elements such as filesrc, videodecoder, tee,…

Another choice is to use some software encoders.

What exactly did you mean by “software encoder”? Do you mean something like the x265enc element? If so, that would require switching my pipeline from GPU to CPU, which would significantly affect my real-time performance.

Then I’ll need to modify my decoder and parser depending on the input, which involves quite a lot of additional coding.

Yes. The hardware limitation should be considered.

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks.