Introduction

Many of today's applications and systems are being monitored to help address performance and network issues. In order to help address the issues that come up, the monitoring tools can launch a data collector to gather important logs and trace information when the monitoring agent detects a problem. In this article, we provide a way to use ITM, a health and performance monitoring tool, to launch the IBM Support Assistant data collector to gather log data when it encounters an error situation. In addition, the results are transferred to an FTP server.

In order to do this, the example in the article uses the following software:

• IBM Support Assistant Lite 1.2.4
• IBM Tivoli Monitoring 6.1
• IBM Tivoli Composite Application Manager for WebSphere

Setup

In our environment, we installed IBM Tivoli Composite Application Manager for WebSphere agent on the WebSphere Application Server we're interested in gathering data for. We configured the agent to communicate with the IBM Tivoli Monitoring Tivoli Enterprise Portal. In addition, we placed an IBM Support Assistant Lite data collector on the WebSphere Application Server. In order to get a better understanding of what role each component plays, we'll introduce each of the different components in a little more detail.

Figure 1: Overview of the components involved

IBM Support Assistant 1.2.4
The tool can be used to troubleshoot a number of IBM software products. It provides a means to automate the data collection procedures ("Must Gathers") defined by the product groups. The tool handles automatic data collection and basic symptom analysis. The tool assists with reducing the time it takes to collect data and the effort it takes to send the appropriate log information to IBM or other ftp servers.

IBM Tivoli Monitoring 6.1
IBM Tivoli Monitoring provides the capability to monitor systems through a single portal. IBM Tivoli Monitoring provides the capability to automate recovery when the monitored system encounters error conditions.

IBM Tivoli Composite Application Manager for WebSphere
IBM Tivoli Composite Application Manager for WebSphere offers a comprehensive deep dive diagnosis capability into the WebSphere application server environment. When combined with a product-provided ITM agent (TEMA) it allows operations team a single integrated view into the environment via ITM and Tivoli Enterprise Portal.

Back to top

Running the Data Collector

Begin by installing IBM Support Assistant Lite onto the system with your WebSphere Application Server. There are many different data collections that are available. For the purpose of this article, we'll be using the general problem for WebSphere Application Server. Typically, the IBM Support Assistant data collectors require user interactions - most of the data collectors need input from the user such as the product location and detailed information about the error. However, the IBM Support Assistant data collectors allow users to run them silently - which means that other tools (such as IBM Tivoli Monitoring) can launch the data collector tool programmatically.

If the JAVA_HOME environment isn't set, you'll need to make sure to set it to the location of your Java™ Runtime Environment.

Windows: set JAVA_HOME="C:/Program Files/IBM/Java142"
Linux: export JAVA_HOME=/opt/IBM/WebSphere/AppServer/java

In order to run in silent mode, you'll need to run IBM Support Assistant Lite in console mode.

Windows: runISALiteConsole.bat
Linux: runISALiteConsole.sh

Figure 2: Starting the collector

Since the tool requires input, you will need to run the data collector first before any tool can programmatically run the data collector. With the first run, you can record and store the input into a file for use in later runs. (e.h. runISALiteConsole.bat -record filename) One this is launched, all your responses will be stored into the filename specified. The input file will only work for collections that are run on the same system. If you have multiple systems where you want to launch the data collector programmatically, you will need to create a new input file for each system. The other option is to manually open and edit the file (however, you'll want to test the script first to make sure that the values still match up).

Figure 3: Sample response file output

For our script, we're going to run the WebSphere Application Server General Problem Script. This will allow us to collect many of the files needed to debug problems with the WebSphere Application Server. When running, you'll want to select the option to view the WebSphere Application Server scripts and the script is located under the "General" category.

Figure 4: Selecting the WebSphere Application Server general problem script

As you run the script, you'll fill in the required information that the collector needs to know about your environment. When running the script, be sure to select the option to FTP the results to a FTP server of your choice. This will allow you to direct all the collected data from different servers and systems at one consolidated location.

Figure 5: Setting the collector to FTP results to a Server

To run IBM Support Assistant Lite using pre-populated values, you can pass in the file name as the first argument. (e.g. runISALiteConsole.bat filename). You'll want to test your script first to make sure there are no errors before continuing to the next step. Once you're done, you should see the collection zip file in the directory you specified (and on the FTP server if you chose that option).

Figure 6: Files collected with IBM Support Assistant Lite

Back to top

Set up trigger

Ideally, we'd like to be able to collect data right after a problem occurs to easily isolate the problem in the logs (and to make sure that no data is lost). For more complicated problems, one might want to collect the data at intervals to compare the results over time. Since the ISA data collector can get started through a command line, the next natural transition is to have an application, such as IBM Tivoli Monitoring, start a collection automatically. For the article, we'll create a situation that will run a data collection when the WebSphere Application Server stops for any reason. To do this, we want the monitoring agent to perform an action when the server is offline. We will go through the steps on how to configure the agent to run a general data collection against our application server and send the results to our FTP server.

1. In the Tivoli Enterprise Portal, find the Websphere Agent. We want to create the situation for that agent, so right click on the agent. Select the Create new situation icon in the toolbar.
2. In the Create Situation window, type in the name of your situation (we'll use WebSphere_Stop). You should be able to see the situation on the side navigation pane now. In addition, you should be able to specify more details in the Situation Editor.

   Figure 7: Creating a situation

   
   
   
3. In the Select Condition editor, you can specify the attributes to compare. We want to do an attribute comparison for the attribute group Application server status for the attribute item status.
4. At this point, you can specify more details such as the cell/node/server name.
5. The Action tab of the Situation Editor allows you to run a system command. For the system command, you'll want to set it to run the IBM Support Assistant Lite console.

   Figure 8: Configure system command

   
   
   
6. Click OK to save. The situation should begin automatically.

Testing trigger

When the WebSphere Application Server is offline, IBM Tivoli Monitoring 6.1 will display that the server is offline. At this point, you should be able to see the data collection zip file on the local application server.

Figure 9: Testing the situation

Conclusion

In this article, we have shown how to use IBM Support Assistant with IBM Tivoli Monitoring and IBM Tivoli Composite Application Manager for WebSphere to perform data collections when certain error situations occur. By using different situations and other data collections, you can configure your environment to collect data for more complicated situations. This will allow you to quickly capture data and allow for a faster resolution to problems that the system may encounter. By using both monitoring tools and data collection tools simultaneously, this should help to cut down on the time it takes to identify and resolve problems.

Resources

Learn

• Automate data collection for problem determination series: Other articles in this series describe what the to how to customize it with Ant scripts, and explain the tool's symptom analysis functions.
  
  
• ITCAM for WebSphere Information Center: Visit this site for more information on IBM Tivoli Composite Application Manager for WebSphere.
  
  
• developerWorks® Tivoli zone: Visit this site for articles, tutorials, news, and other resources on Tivoli technologies.
  
  

Get products and technologies

• Download IBM Support Assistant Lite: Grab the latest version of IBM Support Assistant Lite. The user guide is also available.
  
  

Discuss

• Check out developerWorks blogs and get involved in the developerWorks community.
  
  

About the authors

Belinda Chang is a Software Engineer for IBM in Research Triangle Park, North Carolina, in the Software Group Advanced Design and Technology team. She received a Bachelors degree in Computer Science from Carnegie Mellon University. You can contact Belinda at belchang@us.ibm.com.

Roger Meli is a Software Architect for IBM Tivoli in Research Triange Park, North Carolina. He has a Bachelor's degree from Warwick University, England and a Ph.D. from Cambridge University, England. He can be contacted at rmmeli@us.ibm.com.

Lorne Long is a Software Architect for IBM Tivoli in Research Triange Park, North Carolina. He can be contacted at longlo@us.ibm.com.

Lori Paratore works for IBM Federated Integration Test Team in RTP, NC. Lori received her Bachelor's degree in Computer Science from Villanova University. Lori specializes in cross - product integration and support issues.

Brad Topol is a Senior Technical Staff Member with the Software Group Advanced Design and Technology team at IBM in Research Triangle Park, North Carolina. He received a Ph.D. degree in Computer Science from the Georgia Institute of Technology in 1998. Over the years, Brad has been actively involved in advanced technology projects in the areas of autonomic computing, Web services, grid computing, Web content transformation, and aspect-oriented programming.

Comments (Undergoing maintenance)

Back to top

Trademarks  |  My developerWorks terms and conditions

Table of contents

• Introduction
• Setup
• Running the Data Collector
• Set up trigger
• Testing trigger
• Conclusion
• Resources
• About the authors
• Comments

Next steps from IBM

• Try: WebSphere Application Server
• Webcast: Using Rational Application Developer with WebSphere Application Server to increase productivity
• Briefing: Building SOA solutions and managing the service lifecycle
• Buy: WebSphere Application Server

My developerWorks community

Dig deeper into Tivoli on developerWorks

Rate a product. Write a review.

Write a review for Rational products today!

Special offers