Hello, I got new Dell PowerEdge r730 server with factory installed Nvidia Tesla m40 in it. I installed ESXi hypervisor and Debian 9 GNU/Linux OS on top of it. I also installed proprietary drivers from the non-free package. After that I noticed the nvidia-smi (the tool for monitoring nvidia cards) error message:
“Unable to determine the device handle for GPU 0000:13:00.0: Unable to communicate with GPU because it is insufficiently powered. This may be because not all required external power cables are attached, or the attached cables are not seated properly.”
Therefore, graphic card is not working properly. The graphic card has just one 8 pin power connector, which is connected to the power supply. ESXi and Debian see the graphic card as an installed device. I tried multiple versions of Nvidia drivers, however, it did not help to get rid of the error message.
Could somebody suggest solution for dealing with power supply error from nvidia-smi?