NVdCF tracker issue with 2 T4

• Hardware Platform Nvidia Tesla T4
• DeepStream Version 5.1

• TensorRT Version 7.2.2.3
• NVIDIA GPU Driver Version (valid for GPU only) 460.32…03
• Issue Type( questions, new requirements, bugs) question

Hi, while running deepstream with NVDCF tracker with 2 T4 , i am getting the error-
“GPUassert: invalid device ordinal src/modules/NvDCF/NvDCF.cpp 1250”

No idea with this as it is running fine when using KLT tracker.

Please suggest…
Thanks.

KLT tracker does not use GPU.
Can you run the sample without nvdcf tracker?

Hi,
As i already said, it works with KLT tracker.
Also without nvdcf tracker it works

Can you specify how we could repro your issue with nvdcf tracker?

I am using 16 cameras with 2T4. OS is Redhat8.
For 8 cameras i have provided gpu-id=0 and rest gpu-id=1

Also checked “nvbuf-memory-type=0/3”.

But when i used KLT, then i have to make nvbuf-memory-type=0 and it runs.
With NvDCF, nothing works and same error for GPUassert.

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

Can you share the configuration run failed?