hcc-chase-shor-c4705.unl.edu, hcc-chase-shor-c4709.unl.edu, hcc-chase-shor-c4715.unl.edu, hcc-chase-shor-c4719.unl.edu GPU's offline
8x 3090 GPU nodes at UNL have been losing GPUs over the past few months. I've set the power level to low and am monitoring the GPU counts now.
Some of the nodes have permanently lost GPUs that do not come up after reboots, I moved the cards across the PCI slots and fixed a few of them. Planning to test each GPU individually to determine if any need RMA.