A remap or page-retirement attempt failed — treat as failing memory and drain.
Xid 64 means the GPU tried to remap a row (or retire a page) and the operation failed. Unlike Xid 63, this is not a successful recovery — the device could not protect itself, so the memory should be considered unreliable.
NVRM: Xid (PCI:0000:65:00): 64, pid='N/A', Row Remapper: Failed to remap rowDrain the node. The GPU's self-protection has failed; plan an RMA if it persists after a reset.
The datacenter decision path for hardware-class Xids: cordon → drain → reset → RMA on recurrence. Cordon to stop new work landing, drain running jobs, reset the GPU to clear or remap the fault, and RMA if it comes back.
Real signal. Rare, and consequential — must not be absorbed.
That judgment — absorb the noise, surface the few that are real — is what Plexus does by default on your existing DCGM / Prometheus / Thanos. See why a GPU fleet throws 50 alerts a week and only two matter.
Severity here is an operations judgment for datacenter fleets and depends on driver and GPU generation — always cross-check NVIDIA’s official Xid table.