Troubleshooting guide

I. Introduction

This Myrinet-2000 Installation and Troubleshooting Guide describes the hardware and

software installation procedures for a Myrinet-2000 cluster. Section II summarizes the

required hardware, and Section III provides detailed installation instructions for each

hardware component. Sections IV, V, VI, and VII address the software installation of

MX, GM-2, or GM-1, and Section VIII describes the testing and validation of the

Myrinet cluster. Appendices A and B provide diagnostics for determining if a problem

is hardware- or software-related, as well as procedures for isolating the source of a

hardware failure. Appendix C details troubleshooting procedures for performance

abnormalities.

II. What Hardware Is Required?

A Myrinet-2000 network consists of the following hardware components connected to

your host computers.

Myrinet-2000 PCI-X or PCI64 Network Interface Cards (NICs)

Myrinet-2000 M3-E* or M3-CLOS-ENCL-*/M3-SPINE-ENCL-* Switch(es)

Myrinet-2000 Fiber cables

The basic requirements are:

NICs: one per host

Switches: at least one port per host (more required for clusters larger than

128 hosts when using M3-E* switches, or if using PCI-X NICs

with multiple ports).

Cables: one per port on each NIC (connecting to a port on a switch), plus

any required between switches.

Detailed Myrinet-2000 product specifications are available:

http://www.myri.com/myrinet/product_list.html

III. Hardware Installation

Upon receipt of the Myrinet hardware, we recommend reading the following documents.

For Myrinet-2000 M3-E* switches, please read: