tt_pcie_link_alive

<< Home | << Metrics

Name

Prometheus Metric Name

tt_pcie_link_alive

Metric Path (tt-telemetry)

Schema:

{hostname}/tray{tray}/chip{chip}/pcie/tt_pcie_link_alive

Example path:

bh-glx-c09u02/tray1/chip2/pcie/tt_pcie_link_alive

Description

Indicates whether the chip’s PCIe link responds to a host read. Each collection cycle the telemetry server asks UMD’s hang detector to read a BAR register that the chip is guaranteed to never legitimately hold as 0xFFFFFFFF. If the read returns 0xFFFFFFFF, the PCIe link has silently dropped and any subsequent reads will also return the all-ones fault signature — recovery usually requires a board reset.

This metric is only created for MMIO-capable chips on Wormhole and Blackhole architectures, since those are the devices for which UMD provides a PCIe hang detector. Remote chips and other architectures are skipped (no metric is emitted).

Values

Type: Boolean

Units: None

Allowable values:

• True (1): The chip responded normally to the PCIe probe read.

• False (0): The chip returned the 0xFFFFFFFF fault signature; the PCIe link is hung.

Prometheus Labels

Label Name	Value
hostname	The host from which the metric was collected.
hall	The datacenter hall where the host is located. Sourced from the Factory System Descriptor (FSD).
aisle	The datacenter aisle where the host is located. Sourced from the Factory System Descriptor (FSD).
rack	The rack number where the host is located. Sourced from the Factory System Descriptor (FSD).
shelf_u	The shelf U position in the rack where the host is located. Sourced from the Factory System Descriptor (FSD).
tray	The tray (UBB) that the device is located on.
chip	The ASIC location within the tray.