CentOS, M2090s on a Dell 720

Graphics / Linux Linux
1

This was running. I’ve got two M2090’s in an R720, running CentOS 6.5 I just updated the system, and the GPUs are inaccessible. enum_gpu, which we use to wake them up, fails, presumably having found 0, as does my user’s program.

As a last resort, I yum removed kmod-nvidia and d/l the proprietary package from NVidia, NVIDIA-Linux-x86_64-325.15.run. It seems to build fine, I reboot, and lsmod shows it in, while modinfo tells me that it looks like what I just built. /dev/nvidia* are there, and all are a+w,a+r… but no joy.

Clues for the poor?

    mark
2

Following myself up - note that this is a headless server. I see, in the nvidia-installer.log, that
NVRM: loading NVIDIA UNIX x86_64 Kernel Module 325.15 Wed Jul 31 18:50:56 PDT 2013
nvidia 0000:05:00.0: irq 113 for MSI/MSI-X
NVRM: RmInitAdapter failed! (0x25:0x48:1157)
NVRM: rm_init_adapter(0) failed
NVRM: VM: nvidia_exit_module:1127: 0xffff882052e914c0, 20 page(s), count = 1, flags = 0x00010011, key = 0x2051cb9000, page_table = 0xffff882051025388

but don’t know if that has to do with no video display.

   mark
3

Following myself up - note that this is a headless server. I see, in the nvidia-installer.log, that
NVRM: loading NVIDIA UNIX x86_64 Kernel Module 325.15 Wed Jul 31 18:50:56 PDT 2013
nvidia 0000:05:00.0: irq 113 for MSI/MSI-X
NVRM: RmInitAdapter failed! (0x25:0x48:1157)
NVRM: rm_init_adapter(0) failed
NVRM: VM: nvidia_exit_module:1127: 0xffff882052e914c0, 20 page(s), count = 1, flags = 0x00010011, key = 0x2051cb9000, page_table = 0xffff882051025388

but don’t know if that has to do with no video display.

   mark