Readme file for IBM® Spectrum LSF 10.1 Fix 601277

Abstract

P104739. The following fix:

·        Prevents the RUNLIMIT from being ignored when a client host is configured in lsb.hosts file.

·        Treats total gmem for GPU of preemptable jobs as available during preemption scheduling.

Description

Readme documentation for IBM® Spectrum LSF 10.1 Fix 601277 including installation-related instructions, prerequisites and co-requisites, and list of fixes.

This fix addresses the following issue:

·        When you configure a client host in the lsb.hosts file under the host section, the RUNLIMIT is ignored after the mbatchd daemon reconfigures or restarts.

·        Currently, LSF does not collect gmem usage of a job. During preemption scheduling, gmem for GPU used by preemptable jobs cannot be released to schedule high priority jobs. After this fix, LSF treats total gmem for GPU of a preemptable job as available during preemption scheduling, then high priority jobs might preempt low-priority jobs successfully. The gmem is not preemptable resource and is GPU-based, not host-based. Even with this fix, it still cannot have same behavior with the resources that can be configured in PREEMPTABLE_RESOURCES. LSF cannot trigger preemption by high priority job that is lacking only gmem.

Readme file for: IBM® Spectrum LSF

Product or component release: 10.1

Update name: Fix 601277

Fix ID: LSF-10.1-build601277

date: 19 August 2022

Contents

1. List of fixes

2. Download location

3. Product or components affected

4. System requirements

5. Installation and configuration

6. List of files

7. Product notifications

8. Copyright and trademark information

 

1. List of fixes

P104739

 

2. Download locations

Download fix 601277 from the following location: https://www.ibm.com/support/fixcentral

 

3. Product or components affected

Affected product or components include:

LSF/bjobs

LSF/mbatchd

LSF/mbschd

LSF/ebrokerd

 

4. System requirements

linux2.6-glibc2.3-x86_64

linux3.10-glibc2.17-x86_64

 

5. Installation and configuration

Before you install

LSF_TOP=Full path to the top-level installation directory of LSF.

1.      You must have LSF 10.1 Fix Pack 12 or above installed before installing this fix. Download this fix pack from IBM Fix Central (https://www.ibm.com/support/fixcentral) and search for build600488. Contact IBM LSF Support if you have any questions or problems with installing Fix Pack 12.

2.      Starting in IBM Spectrum LSF version 10.1 Fix Pack 13, the default values of the following three GPU parameters are changed to:

LSF_GPU_AUTOCONFIG=Y

LSB_GPU_NEW_SYNTAX=extend

LSF_GPU_RESOURCE_IGNORE=Y

If you have Fix Pack 13 installed, no further action is needed to set these parameters. If you have IBM Spectrum LSF version 10.1 Fix Pack 12 or earlier, consider explicitly configuring the same values to these three parameters.

If you want to keep the former GPU behavior, and any one of the three parameters are missing in the lsf.conf configuration file, you must explicitly configure the following default settings that are defined in Fix Pack 12 or earlier:

LSF_GPU_AUTOCONFIG=N

LSB_GPU_NEW_SYNTAX=N

LSF_GPU_RESOURCE_IGNORE=N

3.      Log on to the LSF management host as root.

4.      Set the LSF cluster environment:

- For csh or tcsh: % source LSF_TOP/conf/cshrc.lsf

- For sh, ksh, or bash: $ . LSF_TOP/conf/profile.lsf

Installation steps

1.      Go to the install directory of the fix: cd $LSF_ENVDIR/../10.1/install/

2.      Copy the fix file to the install directory:  $LSF_ENVDIR/../10.1/install/

3.      Run patchinstall: ./patchinstall <fix>

After you install

1.      Log on to the LSF management host as the LSF cluster primary administrator and set the LSF cluster environment.

2.      Run badmin mbdrestart

Uninstallation

1.      Log on to the LSF management host as root and set the LSF cluster environment.

2.      Go to the install directory of the fix: cd $LSF_ENVDIR/../10.1/install/

3.      Run ./patchinstall -r <fix>

4.      Log on to the LSF management host as the LSF cluster primary administrator and set the LSF cluster environment.

5.      Run badmin mbdrestart

 

6. List of files

bjobs

mbatchd

mbschd

ebrokerd

 

7. Product notifications

To receive information about product solution and fix updates automatically, subscribe to product notifications on the My notifications page (www.ibm.com/support/mynotifications) on the IBM Support website (support.ibm.com). You can edit your subscription settings to choose the types of information you want to get notification about, for example, security bulletins, fixes, troubleshooting, and product enhancements or documentation changes.

 

8. Copyright and Trademark Information

©Copyright IBM Corporation 2022

U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.

IBM®, the IBM logo, and ibm.com® are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at "Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml.