Readme file for IBM Watson® Machine Learning Accelerator Interim Fix 547571  

 

Readme file for: IBM Watson® Machine Learning Accelerator
Product/Component Release: 1.2.1
Fix ID: dli-1.2.3-build547571-wmla

Publication date: May 22, 2020

 

Interim Fix 547571 introduces enhancements and resolves the following issues in the IBM Spectrum

Conductor Deep Learning Impact 1.2.3 component in IBM Watson Machine Learning Accelerator 1.2.1:

·       Fixes the issue where access is denied to a folder that contains a model and the permission setting on that folder is 700.

·       Supports setting permission for the sub directories in the DLI_RESULT_FS shared directory to 750.

 

Contents

1.     Download location 

2.     Products or components affected

3.     Installation and configuration

4.     Uninstallation

5.     List of files

6.     Product notifications 

7.     Copyright and trademark information

 

1.   Download location

Download interim fix 547571 (dli-1.2.3-build547571-wmla) from the following location http://www.ibm.com/eserver/support/fixes/.

 

2.   Products or components affected

Component Name, Platform, Fix ID:

DLPD

Linux x86_64, Linux ppc64le

dli-1.2.3-build547571-wmla

 

3.   Installation and configuration

3.1 Before installation

Before installing the interim fix, complete the following steps to prepare your environment.

 

1.     Upgrade to IBM Spectrum Conductor 2.4.1 by applying interim fix 536919

2.     Log on to the management host as a cluster administrator (CLUSTERADMIN) and source the environment according to your shell environment. 

For sh, ksh or bash:

> . $EGO_TOP/profile.platform

> export DLI_SHARED_FS=$DLI_SHARED_FS

> export CLUSTERADMIN=$CLUSTERADMIN

where EGO_TOP is IBM Spectrum Conductor Deep Learning Impact installation path, and DLI_SHARED_FS, CLUSTERADMIN must be the same as the IBM Spectrum Conductor Deep Learning Impact installation setting.

 

For csh or tcsh:

> source $EGO_TOP/cshrc.platform

> setenv DLI_SHARED_FS $DLI_SHARED_FS

> setenv CLUSTERADMIN $CLUSTERADMIN

where EGO_TOP is IBM Spectrum Conductor Deep Learning Impact installation path, and DLI_SHARED_FS, CLUSTERADMIN must be the same as the IBM Spectrum Conductor Deep Learning Impact installation setting.

 

3.     Create a backup directory, for example /build547571_backup, and back up the following files:

> cp $DLI_SHARED_FS/tools/dataset/main.py /build547571_backup

> cp $DLI_SHARED_FS/tools/dataset/utils/common.py /build547571_backup

> cp $DLI_SHARED_FS/conf/spark-env.sh /build547571_backup

> cp $DLI_SHARED_FS/conf/access_control_util.sh /build547571_backup

> cp $DLI_SHARED_FS/tools/dl_plugins/common_wrapper.sh /build547571_backup

> cp $DLI_SHARED_FS/tools/dl_plugins/disttensorflow_wrapper.sh /build547571_backup

> cp $DLI_SHARED_FS/tools/dl_plugins/PowerAICaffeIBM_wrapper.sh /build547571_backup

> cp $DLI_SHARED_FS/tools/dl_plugins/pytorch_wrapper.sh /build547571_backup

> cp $DLI_SHARED_FS/tools/dl_plugins/tensorflow_wrapper.sh /build547571_backup

 

3.2 Installation steps

After successfully upgrading to IBM Spectrum Conductor 2.4.1 and installing interim fix 536919, apply interim fix 547571 by completing the following steps.

 

1.      Log on to the management host as the cluster administrator (CLUSTERADMIN) and source the environment.

For sh, ksh or bash:

        > . $EGO_TOP/profile.platform

For csh or tcsh:

        > source $EGO_TOP/cshrc.platform

 

2.      Stop the following services:

> egosh service stop dlpd

 

3.      As the cluster administrator (CLUSTERADMIN), on management host, download the packages to a directory. For example, packages can be downloaded to the /dlifixes directory.

 

4.      On management host, run the egoinstallfixes command to install cluster jars.

For Linux x86_64:

> egoinstallfixes /dlifixes/dlicore-1.2.3.0_x86_64_build547571.tar.gz

 

For Linux ppc64le:

> egoinstallfixes /dlifixes/dlicore-1.2.3.0_ppc64le_build547571.tar.gz

 

NOTE: Running the egoinstallfixes command automatically backs up the current binary files to a fix backup directory for recovery purposes. Do not delete this backup directory; you will need it if you want to recover the original files. For more information on using this command, see the egoinstallfixes command reference.

 

5.      Run the pversions command to verify the installation:

> pversions -b 547571

   

6.   Log on to the management host as a cluster administrator (CLUSTERADMIN), download the dli-1.2.3.0_build547571_share.tar.gz package and extract its contents to the top-level $DLI_SHARED_FS directory:

> tar zoxf dli-1.2.3.0_build547571_share.tar.gz -C $DLI_SHARED_FS

 

7.   Ensure that the dli-1.2.3.0_build547571_share.tar.gz patch files are set with correct permissions:

> chmod 755 -R $DLI_SHARED_FS/tools/dataset

> chmod 755 $DLI_SHARED_FS/conf/access_control_util.sh

> chmod 775 $DLI_SHARED_FS/conf/spark-env.sh

 

8. Log on to the management host as a cluster administrator (CLUSTERADMIN) and update the dlpd configuration file.

 

a. Open the dlpd configuration file, run the following command:

> vi $EGO_CONFDIR/../../dli/conf/dlpd/dlpd.conf

 

b. Edit the dlpd configuration file by adding the following line to define a new parameter:

"DLI_RESULT_FS_SUBDIR_PERMISSION": "750"

Note: If this value is not specified, permissions are set to 700 by default.

 

c. Save your changes and close the file.

Important: The permission settings will be applied to new directories created under the DLI_RESULT_FS shared directory. If you have existing directories, you must run the following command to apply permission settings to existing directories:

> chmod 750 -R $DLI_RESULT_FS/*

 

9.      Start the following services:

> egosh service start dlpd

 

10.   Log out of the cluster management console, clear your browser cache, and relogin.

 

4.   Uninstallation

If required, follow the instructions in this section to uninstall this interim fix on hosts in your cluster.

 

1.      Log in to the management host as a cluster administrator (CLUSTERADMIN) and source the environment.

 

2.      Stop the following services:

> egosh service stop dlpd

 

3.      Log on to each management host in the cluster and roll back this interim fix:

> egoinstallfixes -r 547571

 

4.      Restore the files that you backed up during installation under $DLI_SHARED_FS.

> cp /build547571_backup/main.py $DLI_SHARED_FS/tools/dataset

> cp /build547571_backup/common.py $DLI_SHARED_FS/tools/dataset/utils

> cp /build547571_backup/spark-env.sh $DLI_SHARED_FS/conf

> cp /build547571_backup/access_control_util.sh $DLI_SHARED_FS/conf

> cp /build547571_backup/common_wrapper.sh $DLI_SHARED_FS/tools/dl_plugins

> cp /build547571_backup/disttensorflow_wrapper.sh $DLI_SHARED_FS/tools/dl_plugins

> cp /build547571_backup/PowerAICaffeIBM_wrapper.sh $DLI_SHARED_FS/tools/dl_plugins

> cp /build547571_backup/pytorch_wrapper.sh $DLI_SHARED_FS/tools/dl_plugins

> cp /build547571_backup/tensorflow_wrapper.sh $DLI_SHARED_FS/tools/dl_plugins

 

5.      Start the following services:

> egosh service start dlpd

 

5.   List of files 

 

dlicore-1.2.3.0_<arch>_build547571.tar.gz:

$EGO_TOP/dli/1.2.3/dlpd/lib/cws_dl-common-1.2.3.jar

$EGO_TOP/dli/1.2.3/dlpd/lib/cws_dl-core-1.2.3.jar

 

dli-1.2.3.0_build547571_share.tar.gz:

$DLI_SHARED_FS/tools/dataset/main.py

$DLI_SHARED_FS/tools/dataset/utils/common.py

$DLI_SHARED_FS/conf/spark-env.sh

$DLI_SHARED_FS/conf/access_control_util.sh

$DLI_SHARED_FS/tools/dl_plugins/common_wrapper.sh

$DLI_SHARED_FS/tools/dl_plugins/disttensorflow_wrapper.sh

$DLI_SHARED_FS/tools/dl_plugins/PowerAICaffeIBM_wrapper.sh

$DLI_SHARED_FS/tools/dl_plugins/pytorch_wrapper.sh

$DLI_SHARED_FS/tools/dl_plugins/tensorflow_wrapper.sh

 

6.   Product notifications

To receive information about product solution and patch updates automatically, subscribe to product notifications on the My Notifications page http://www.ibm.com/support/mynotifications/ on the IBM Support website (http://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. 

 

7.   Copyright and trademark information 

© Copyright IBM Corporation 2020

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