Yes, I solved it. For me (on Ubuntu) I had to use nvidia-smi as root (not sudo, but as root). then type nvidia-smi --help and the command for establishing persistence will be there <-pm 1>.
$ sudo -i
# nvidia-smi -pm 1
# exit
$
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Nvidia-persistenced: Failed to query NVIDIA devices | 8 | 10293 | August 18, 2023 | |
| Failed to start nvidia-persistenced.service | 0 | 3036 | May 28, 2023 | |
| Setting up nvidia-persistenced | 12 | 46556 | July 19, 2020 | |
| Troubles with starting NVIDIA Persistence Daemon service (Ubuntu 16.04) | 0 | 1174 | January 25, 2020 | |
| After installing CUDA 9.0 in POWER9(RHEL7), nvidia-smi shows Unknown Error in Memory_Usage column. | 18 | 3134 | June 8, 2018 | |
| Failed to initialize NVML: Unknown Error when running nvidia-smi on Docker container | 2 | 10469 | October 18, 2020 | |
| NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running. | 2 | 5668 | August 16, 2019 | |
| Cuda Installation on Ubuntu 18.04 Failing | 8 | 2792 | March 26, 2020 | |
| Headless K80 system: NVIDIA drivers (/dev/nvidia) not created at install time | 12 | 2311 | March 21, 2021 | |
| Nvidia-persistenced fails to start if user option is set to non-root user | 4 | 6677 | October 12, 2021 |