cuda

[elipool]

so its telling me that it can not find my cuda ? what is that

[sigstart]

Can confirm that nanominer-linux 3.6.5 and 3.6.6 ask for libcuda.so on AMD rigs. Rolled back to nanominer 3.6.4 and everything worked again.

[Tupilary]

Running a XMR only setup: the last two versions of nanominer-linux fail to start (3.6.6 & 3.6.7). Version 3.6.5 starts with the following error message, but it continues to process: 2022-Jun-30 15:53:56: CUDA call failed: no CUDA-capable device is detected

The later versions terminate immediately with: terminate called after throwing an instance of 'std::runtime_error' what(): no CUDA-capable device is detected (cudaErrorNoDevice)

[fri-K]

/opt/nanominer/default/nanominer --version
terminate called after throwing an instance of 'std::runtime_error'
  what():  no CUDA-capable device is detected (cudaErrorNoDevice)

/opt/nanominer/default -> /opt/nanominer/nanominer-linux-3.6.7/

I have nvidia drivers installed, but at this moment there is no nvidia card present in the rig

[Tupilary]

Running a XMR only setup: the last two versions of nanominer-linux fail to start (3.6.6 & 3.6.7). Version 3.6.5 starts with the following error message, but it continues to process: 2022-Jun-30 15:53:56: CUDA call failed: no CUDA-capable device is detected

The later versions terminate immediately with: terminate called after throwing an instance of 'std::runtime_error' what(): no CUDA-capable device is detected (cudaErrorNoDevice)

I can confirm that nanominer-linux-3.6.8 has fixed "cudaErrorNoDevice" for me. I have installed from scratch, and allowed the autoUpdate process to upgrade an instance.

Thank you for the fix.