ibdiagnet InfiniBand Fabric Diagnostic Tool User Manual v2.8.0

ibdiagnet Functionality

Functionality

Description

Fabric Discovery

Sweeps the InfiniBand fabric and collects information from the following InfiniBand devices:

  • Switches

  • HCAs

  • Routers

  • Aggregation Nodes

  • Gateways

Duplicated GUIDs Detection

Checks and reports duplicated Node and Port GUIDs in the fabric.

Duplicate Node Description Detection

Checks and warns regarding duplicated node description of switches or HCAs.

Alias GUIDs Check

Performs alias GUID checks (Relevant for ConnectX-3 devices only)

LIDs Check

Performs correct LID assignment and duplicated LID check for InfiniBand devices.

Links in INIT State and Unresponsive Nodes Detection

Reports links in INIT logical state. Additionally, it reports unresponsive devices and Direct Route to such devices.

Split Cables Support

Reports split cables in ibdiagnet dumps and logs.

Counters Fetch

Fetches various counters from InfiniBand devices, including standard and extended port counters, diagnostic counters, phy counters, etc.

Error Counters Check

Checks errors counters crossing thresholds between counter snapshots.

Routing Fetch and Checks

Performs correctness check of the switch routing tables as well as checking of credit-loop free routing.

Link Width and Speed Checks

Checks that fabric links are operating at maximum supported speed and width.

Dumping Virtualization Information

Dumps virtualization information from channel adaptors.

Dumping SHARP Trees Structures and SHARP Counters

Dumping SHIELD Configuration and Counters

Topology Matching

Performs matching of fabric topology with previously stored one.

Fast Discovery

Avoids rediscovering the fabric by using previously cached fabric data.

Support IB Security

Using MKEY, VSKEY, AMKEY and CC KEY for dumping fabric configuration.

Partition Checks

Dumps and validates HCA and switch partition tables.

BER Test

Reports links with high Bit Error Rate (BER).

Dump PCI Data from Servers

Dump Cable Info

Dump PHY Info

© Copyright 2023, NVIDIA. Last updated on May 23, 2023.