Skip to navigation
ReferenceCheck Catalog

Check Catalog

View as Markdown

Every check this build registers, grouped by hardware domain. Each page lists a check’s ID, the signals it reads, its thresholds, and the rules that decide its severity.

DomainChecksCovers
GPU5GPU covers GPU device health: driver-reported events (XID), thermal state, driver status, power, and clock behavior.
NVLink6NVLink covers NVLink interconnect health: replay counters, link errors, and fabric state.
InfiniBand1InfiniBand covers InfiniBand network health: port state, error rates, and link integrity.
Node1Node covers host-wide GPU-system signals that are not bound to any single device — GPU count, driver version, and similar inventory checks.
Xid47Xid covers NVRM Xid fault events logged by the NVIDIA driver to the kernel ring buffer. Xid measurements are collected separately from GPU telemetry and route only to Xid checks.
CPU2CPU covers host processor inventory and availability signals.