IBM Spectrum LSF Explorer V10.2.0 Fix Pack 9 Readme
Abstract
This release includes features and enhancements since the release of IBM Spectrum LSF Explorer 10.2.0.8 on May 31, 2019.
Description
Readme documentation for IBM Spectrum LSF Explorer 10.2.0.9 Release including installation-related instructions, prerequisites and co-requisites.
Readme file for: IBM® Spectrum LSF Explorer
Product/Component Release: 10.2.0.9
Update Name: Fix 534236
Fix ID: explorer-10.2.0.9-build534236
Publication date: 15 Nov 2019
Last modified date: 15 Nov 2019
Contents
- What's new - Enhancements and List of Fixes
- System requirements
- Known issues
- Product notifications
- Download location
- Installation and configuration
- Upgrade from 10.2 to 10.2.0.9
- Uninstallation
- Copyright and trademark information
- What's new - Enhancements and List of Fixes
- Support for Elasticsearch 7.2.1
- Support for Logstash 7.2.1
- Kibana 7.2.1 bundled with installer for use with Elasticsearch
- Box and whisker plots for CPU utilization and Slot utilization
- New data loader to collect host property information in lsf_hostmetrics index. Additional fields for finished jobs are collected in lsf_job_acct index
List of Fixes
- Fix ID
|
- Description
|
- P103214
|
- ES data.path can't be defined in install.config
|
- P103209
|
- Too many events logged in lsfjobstatusloader.log on INFO level
|
- P103207
|
- Data not cached in work/backupdata directory when ES switched to read-only
|
- P103175
|
- The name of vendor-feature is displayed incompletely in the filter on LE 10.2.0.8
|
- P103120
|
- Only 15 records will be displayed no matter how long the time interval is specified
|
- System requirements
Refer to Knowledge Center : https://www.ibm.com/support/knowledgecenter/SS6MQM_10.2.0/release_notes/sys_requirements_10.2.0.9.html
- Known issues
Refer to Knowledge Center : https://www.ibm.com/support/knowledgecenter/SS6MQM_10.2.0/release_notes/rel_notes_known_issues.html
- 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.
- Download location Download from the following location: https://www.ibm.com/support/fixcentral
- Installation and configuration
- In this document,
- COLLECTOR_TOP describes the top-level IBM Spectrum LSF Explorer Data Collector installation directory.
- SERVER_TOP describes the top-level IBM Spectrum LSF Explorer Server installation directory.
- Pre-installation task only for customer-installed Elasticsearch 5.x. *Skip this step if you installed the Elasticsearch included with Explorer or are using Elasticsearch 7.x/6.x
For IBM Spectrum LSF Explorer to work with customer installed Elasticsearch 5.x, you need to configure specific parameters
in the elasticsearch.yml configuration file using the following steps
1. Log on to the Elasticsearch Server.
- 2. Add the following configuration parameters in the elasticsearch.yml file. Depending on your installation, it may be
- found at ES_TOP/config/ or in the /etc/ directory. ES_TOP is the top-level directory in which you installed Elasticsearch.
- script.inline: true
- script.stored: true
- script.search: true
- Note: When adding lines to the elasticsearch.yml file, ensure the new lines have no blank spaces before them.
- Otherwise, Elasticsearch reports errors and fails to start.
- 3. Restart the Elasticsearch service
- 6.1 Installation steps for IBM Spectrum LSF Explorer Server 10.2.0.9
- Log in as root to the host on which you intend to install IBM Spectrum LSF
Explorer Server.
- Extract the package.
For example:
tar -zxvf explorer10.2.0.9_server_linux-x64.tar.gz
- Edit the install.config file to set installation variables.
If you choose to install the packaged Elasticsearch in the IBM Spectrum LSF
Explorer Server package, then also configure INSTALL_ES=Y
If you choose to use your own Elasticsearch, then configure INSTALL_ES=N
and configure ES_HOST and ES_PORT of your Elasticsearch properly.
If you choose to install the packaged Kibana, then also configure INSTALL_KIBANA=Y
and configure KI_HOST and KI_PORT properly.
- Run the ExplorerServerInstaller.sh installation script and finish the installation:
./ExplorerServerInstaller.sh -f install.config
Note: For a silent installation, add the option -silent.
- Check that the Elasticsearch service has started:
service elasticsearch-for-lsf status
- Set the IBM Spectrum LSF Explorer Server environment:
For csh or tcsh: source SERVER_TOP/lsfsuite/ext/cshrc.platform
For sh, ksh, or bash: . SERVER_TOP/lsfsuite/ext/profile.platform
- Start the IBM Spectrum LSF Explorer Server.
pmcadmin start
- Check that the IBM Spectrum LSF Explorer Server services have started
pmcadmin list
- Log in to the IBM Spectrum LSF Explorer Server:
Open a web browser and access the IBM Spectrum LSF Explorer Server by
entering the host name or IP address and port number of the Server.
The default port number is 8080.
For example, if your IBM Spectrum LSF Explorer Server IP address is
198.51.100.0, enter in your web browser as the URL:
http://198.51.100.0:8080
The Login screen for the IBM Spectrum LSF Explorer Console is displayed.
- Use the LSF administrator name and password to log in. The default
dashboard Cluster Overview is displayed
- Select Dashboards to view the out-of-box reports.
- To create a new chart, select Charts and click the New button.
- To create a new dashboard, select Dashboards and click the New button
- Check for errors in the log files:
The log files are located in SERVER_TOP:
cd SERVER_TOP/lsfsuite/ext/explorer/server/logs
- 6.2 Installation steps for IBM Spectrum LSF Explorer Data Collector 10.2.0.9 Ensure IBM Spectrum LSF Explorer Server is running. IBM Spectrum LSF
Explorer Server must be started before IBM Spectrum LSF Explorer Data
Collector services.
- Log in to the LSF host on which you intend to install IBM Spectrum LSF
Explorer Data Collector as the primary LSF administrator.
Note: Log in as root if you want to update the rc.boot scripts.
- Extract the package.
For example:
tar -zxvf explorer10.2.0.9_node_linux-x64.tar.gz
- Go to the extracted directory and edit the install.config file.
Configure all required parameters by uncommenting the line for the related parameter.
If you choose to enable the scaled IO statistics feature, configure the
GPFS_HOME and GPFS_DEVICE parameters.
If you choose to enable the energy accounting feature, configure the
IPMI_HOME and LOGSTASH_HOST parameters.
- Run the ExplorerNodeInstaller.sh installation script.
./ExplorerNodeInstaller.sh -f install.config
Note: For a silent installation, add the option -silent.
- Set the environment to start the data collectors, where COLLECTOR_TOP is the
directory in which you installed IBM Spectrum LSF Explorer Data Collector.
For csh or tcsh:
source COLLECTOR_TOP/lsfsuite/ext/perf/conf/cshrc.perf
For sh, ksh, or bash:
. COLLECTOR_TOP/lsfsuite/ext/perf/conf/profile.perf
Note: You might notice an error when you set the environment on the Red Hat
Enterprise Linux 6.x x_64 platform, such as:
-bash: /lib/libc.so.6: No such file or directory
-bash: [: : integer expression expected
Cannot get binary type
To avoid this error, after you set the environment, install the related files by
running the following command:
yum install ld-linux.so.2
- Optional: Enable data collectors in your cluster to collect data for the FLEXlm-Managed Licenses Dashboard.
Note: The data collector for the FLEXlm-Managed Licenses Dashboard is not supported on Linux for Power System Servers 8 Little Endian.
a. Get the FLEXlm utility files lmutil and lmgrd from your license server.
b. Copy the lmgrd and lmutil files from the license server location to the directory NODE_TOP/license/7.0/os_type/etc, where os_type is the operating system type.
c. Assign execute permissions to the lmgrd and lmutil files.
d. Edit the FLEXlm license usage data loader configuration file
COLLECTOR_TOP/conf/dataloader/flexlicusage.properties and change the
ServerList (or FileName) parameter to point to the FLEXlm license servers.
For more information, see the description in the flexlicusage.properties file.
e. Enable the FLEXlm license usage data loader in your cluster by editing the file COLLECTOR_TOP/conf/plc/plc_license_explorer.xml and setting the attribute
Enable="true" in the <DataLoader Name="flexlicusageloader" .../>
element.
<DataLoader Name="flexlicusageloader" Interval="300" Enable="true" LoadXML="dataloader/flexlicusage.xml”/>
- Start the data collector plc service.
perfadmin start all
- Check that the data loaders are running.
plcclient.sh -s
- Check for errors in the data loader log files in the directory
$PERF_LOGDIR/dataloader
- Upgrade from 10.2 to 10.2.0.9
(pre-step) Backup Elasticsearch data
Explorer 10.2.0.9 bundles Elasticsearch 7.2.1. If you choose to upgrade Elasticsearch to this version, it will do a reindexing of the current indices.
It is strongly recommended to do a data backup before proceeding. Refer to https://www.elastic.co/guide/en/elasticsearch/reference/7.2/modules-snapshots.html
for more details.
1) Log in to the Elasticsearch server as root.
2) Configure the Elasticsearch snapshot repository. For the cluster, the repository
should use same path under share file system for each elasticsearch node.
Append "path.repo: /var/lib/elasticsearch_repo" to elasticsearch.yml
(default configuration file path: /etc/elastic-search/)
3) Restart Elasticsearch
4) Create repository location: es_backup
Use a relative path to "es_backup_location" under "/var/lib/elasticsearch_repo". For example:
curl -XPUT "ES_IP:ES_PORT/_snapshot/es_backup" -H 'Content-Type: application/json' -d'{
"type": "fs",
"settings": {
"location": "es_backup_location",
"include_global_state": true,
"compress": true
}
}'
5) Create a snapshot: es_snapshot
curl -XPUT "ES_IP:ES_PORT/_snapshot/es_backup/es_snapshot?wait_for_completion=true"
7.1 Upgrade from 10.2 to 10.2.0.9 for Explorer Server
1. Set the IBM Spectrum LSF Explorer Server environment:
For csh or tcsh: source SERVER_TOP/lsfsuite/ext/cshrc.platform
For sh, ksh, or bash: . SERVER_TOP/lsfsuite/ext/profile.platform
2. Stop the IBM Spectrum LSF Explorer Server.
pmcadmin stop
3. Install LSF Explorer Server 10.2.0.9 following steps in section 6.1
Note: If you use packaged Elasticsearch 5.4.1 or 6.6.1 in the IBM Spectrum LSF
Explorer and choose to upgrade to Explorer bundled Elasticsearch 7.2.1,
configure INSTALL_ES=Y in install.config and Elasticsearch will be
upgraded automatically during Explorer Server upgrading.
If you use your own Elasticsearch and want to upgrade it to 7.x,
you need to configure INSTALL_ES=N in install.config and upgrade
it manually. Please contact IBM support for help.
4. Set the IBM Spectrum LSF Explorer Server environment:
For csh or tcsh: source SERVER_TOP/lsfsuite/ext/cshrc.platform
For sh, ksh, or bash: . SERVER_TOP/lsfsuite/ext/profile.platform
5. Start the IBM Spectrum LSF Explorer Server.
pmcadmin start
7.2 Upgrade from 10.2 to 10.2.0.9 for Explorer Data Collector
1. Source the environment.
For csh or tcsh:
source COLLECTOR_TOP/lsfsuite/ext/perf/conf/cshrc.perf
For sh, ksh, or bash:
. COLLECTOR_TOP/lsfsuite/ext/perf/conf/profile.perf
2. Stop the data loaders.
perfadmin stop all
3. Install LSF Explorer Data Collector 10.2.0.9 following steps in section 6.2.
4. Source the environment.
For csh or tcsh:
source COLLECTOR_TOP/lsfsuite/ext/perf/conf/cshrc.perf
For sh, ksh, or bash:
. COLLECTOR_TOP/lsfsuite/ext/perf/conf/profile.perf
5. Start the data loaders.
perfadmin start all
- Uninstallation 8.1 Uninstall Explorer Server
Run the ExplorerServerUninstall.sh uninstallation script.
. SERVER_TOP/lsfsuite/ext/explorer/server/bin/ExplorerServerUninstall.sh
8.2 Uninstall Explorer Data Collector
Run the ExplorerNodeUninstall.sh uninstallation script
. COLLECTOR_TOP/lsfsuite/ext/perf/1.2/bin/ExplorerNodeUninstall.sh
- Copyright and trademark information ©Copyright IBM Corporation 2019
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
This product includes some components redistributed from third-party vendors which are provided here for convenience. Licensee may copy and use these components within the confines of the third-party license as identified in the Notices file. IBM provides these third-party components without obligation of support and "AS IS", WITH NO WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTY OF TITLE, NON-INFRINGEMENT OR NON-INTERFERENCE AND THE IMPLIED WARRANTIES AND CONDITIONS OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.