K20Xm driver (352.93) error messages

Hi,

On a HPC cluster running RHEL 6, we have two K20Xm graphics cards per node. The previous graphics driver (352.53) worked without error. We have since upgraded to 352.93 and now get a set of 5 messages in /var/log/messages after each reboot on every node:

kernel: nvidia: module license ‘NVIDIA’ taints kernel.
kernel: nvidia 0000:03:00.0: PCI INT A -> GSI 32 (level, low) -> IRQ 32
kernel: nvidia 0000:83:00.0: PCI INT A -> GSI 64 (level, low) -> IRQ 64
kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module 352.93 Tue Apr 5 18:18:24 PDT 2016
nvidia-persistenced: Started (4234)

6 nodes also have shown the following errors after 1 reboot only (but they show the above errors after every reboot as well):

kernel: NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5
kernel: NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5
kernel: NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5
kernel: NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5
kernel: NVRM: nvidia_frontend_open: minor 1, module->open() failed, error -5

Are there any common causes for each of these issues that I can look for? It seems that the GPUs are running - they accept loads and nvidia-smi itself returns no errors. But the fact that these error messages are coming up every time a node is rebooted is worrisome.