Managing Lustre on a Cray XT System

S-0010-21 - Nov 2008

This Technical Note is provided to describe management of Lustre file system(s) on Cray XT systems running the Cray Linux Environment (CLE) release package 2.1 or later.

Links

Available formats

Adobe Acrobat Reader S-0010-21.pdf
Valid HTML 4.01 html-S-0010-21

Table of contents

Lustre Configuration
    1.1  Lustre File System Documentation
    1.2  Lustre Software Components
    1.3  Lustre Framework
    1.4  Lustre File System Configuration
    1.4.1  Lustre Configuration File
    1.4.2  Modifying Lustre Configuration Parameters
    1.5  Updating the Bootimage
    1.6  Configuring the Lustre Lock Recovery Daemon
Lustre File System Management
    2.1  Storage, Network and Command Information
    2.1.1  Storage Devices for Lustre
    2.1.2  Lustre Networking
    2.1.3  Lustre Commands
    2.1.4  Location of Lustre Kernel Modules and Libraries
    2.1.5  Lustre Layout
    2.2  Confirming Lustre File System Definition Files Using lustre_readiness.sh
    2.3  Configuring Striping on Lustre File Systems
    2.3.1  Configuration and Performance Trade-off for Striping
    2.3.2  Overriding File System Striping Defaults
    2.4  Setting Secondary Group Permissions with group_upcall
    2.5  Lustre System Administration
    2.5.1  Identifying MDS and OSTs
    2.5.2  Checking Lustre Disk Usage
    2.5.3  Starting Lustre
    2.5.4  Stopping Lustre
    2.5.5  Checking the Lustre File System
    2.6  Troubleshooting
    2.6.1  Dumping Lustre Log Files
    2.6.2  Lustre Users Report ENOSPC Errors
    2.6.3  Troubleshooting User Applications on Catamount Nodes
    2.6.4  File System Error Messages
Lustre Failover
    3.1  Lustre Failover for CNL (Deferred implementation)
    3.1.1  Node Types for Failover
    3.2  Lustre Manual Failover
    3.2.1  Configuring Manual Lustre Failover
    3.2.2  Performing Manual Failover
    3.2.3  Monitoring Manual Failover
    3.3  Lustre Automatic Failover for CNL
    3.3.1  Lustre Automatic Failover Database Tables
    3.3.2  Backing Up SDB Table Content
    3.3.3  Using the xtlusfoadmin Command
    3.3.4  System Startup and Shutdown when Using Automatic Lustre Failover
Lustre Failback
    4.1  Lustre Failback (Deferred implementation)
    4.1.1  Failback in Manual and Automatic Failover
List of Tables
List of Figures
List of Examples
List of Procedures

Software Releases this book supports

Product Version Sub Product Release Date
Cray Linux Environment (CLE) 2.1 Nov 2008

Other versions of this book

Publication Number Release Date Supported Software Releases
S-0010-5203 Apr 2015 Cray Linux Environment (CLE) 5.2.UP03, Cray Linux Environment (CLE) 5.2.UP04
S-0010-52 Mar 2014 Cray Linux Environment (CLE) 5.2.UP00, Cray Linux Environment (CLE) 5.2.UP02, Cray Linux Environment (CLE) 5.2.UP01
S-0010-4201 Jul 2013 Cray Linux Environment (CLE) 4.2.UP01, Cray Linux Environment (CLE) 4.2.UP02
S-0010-42 Apr 2013 Cray Linux Environment (CLE) 4.2
S-0010-4101 Dec 2012 Cray Linux Environment (CLE) 4.1.UP01
S-0010-5001 Nov 2012 Cray Linux Environment (CLE) 5.0.UP02, Cray Linux Environment (CLE) 5.0.UP03, Cray Linux Environment (CLE) 5.1.UP00, Cray Linux Environment (CLE) 5.1.UP01
S-0010-4002 Dec 2011 Cray Linux Environment (CLE) 4.0.UP02, Cray Linux Environment (CLE) 4.0.UP03
S-0010-4001 Sep 2011 Cray Linux Environment (CLE) 4.0.UP01
S-0010-40 Jun 2011 Cray Linux Environment (CLE) 4.0
S-0010-31 Jun 2010 Cray Linux Environment (CLE) 3.1, Cray Linux Environment (CLE) 3.1.UP02
S-0010-30 Mar 2010 Cray Linux Environment (CLE) 3.0
S-0010-22 Jul 2009 Cray Linux Environment (CLE) 2.2
S-0010-10 Sep 2008 Knowledge Base 1.0