Configuring synthetic transactions

Synthetic transactions are a form of outside-in monitoring which do not require an agent on the monitored object. Synthetic transactions are performed by a third computer, with the Operations Manager agent installed, called a watcher node. Synthetic transactions are tests that are performed to determine the availability or performance of one of the following items:

  • OLD DB Data Source
  • Process
  • TCP Port
  • UNIX or Linux Log File
  • UNIX or Linux Process
  • Web Application Availability
  • Web Application Transaction
  • Windows Service

You can configure Operations Manager to run tests against these items by configuring one of the management pack templates, available through the Authoring workspace of the Operations Manager console, and shown in Figure 1. The Application Performance Monitoring template, also listed here, is covered later in the chapter.

clip_image002

FIGURE 1 Management pack templates

The tests that form the basis of synthetic transactions are initiated from a computer termed a watcher node. A watcher node can be a management server, a computer or device that hosts an Operations Manager agent. When configuring certain synthetic transactions, you specify which computers will function as watcher nodes, as shown in Figure 2.

clip_image003

FIGURE 2 Watcher nodes

More Info: Watcher Nodes

You can learn more about watcher nodes at http://technet.microsoft.com/en-us/library/hh457584.aspx.

OLE DB data source

You can use the OLE DB data source template to create a synthetic transaction to monitor the performance and availability of any database, not just those running Microsoft’s SQL Server, that you can establish a connection to through OLE DB. It’s possible to create this synthetic transaction even when the computer that hosts the database does not have an Operations Manager agent installed.

To create a synthetic transaction using the OLE DB Data Source Wizard, perform the following steps:

1. Under Management Pack Templates, in the Authoring workspace of the Operations Manager console, click OLE DB Data Source, and click Add Monitoring Wizard on the ribbon.

2. On the Monitoring Type page, select OLE DB Data Source, as shown in Figure 3, and click Next.

clip_image004

FIGURE 3 OLE DB Data Source

3. On the General Properties page, provide a name for the monitor, and select an unsealed management pack in which to store the monitor.

4. On the Connection String page, provide the connection string to the database that you will be connecting to with the monitor. Provide a query that will be used to perform the test and specify a timeout. Figure 4 shows a query against the Operations Manager database.

clip_image005

FIGURE 4 Connection string

5. On the Query Performance page, select the measurements that you want to monitor, the error threshold, and the warning thresholds, as shown in Figure 5.

clip_image006

FIGURE 5 Query performance

6. On the Watcher Nodes page, specify the computers with the Operations Manager agent installed from which the OLE DB test will be performed.

7. Complete the wizard to create the OLE DB synthetic transaction.

More Info: OLE DB Data Source

You can learn more about performing tests against OLE DB data sources at http://technet.microsoft.com/en-us/library/hh457575.aspx.

Process

A process monitoring synthetic transaction allows you to monitor whether a specific process is running on a computer. You can use a process monitoring synthetic transaction to determine the following information:

  • Number of processes running
  • Amount of time that the process has been running
  • Processor utilization of the process
  • Process memory utilization

When configuring a process monitoring synthetic transaction, you can also configure an alert to be raised if the processor or memory utilization exceeds a specific threshold, as shown in Figure 6.

clip_image007

FIGURE 6 Performance Data

More Info: Processes

You can learn more performing tests against processes at http://technet.microsoft.com/en-us/library/hh457551.aspx.

TCP Port

A TCP Port-based synthetic transaction allows you to perform a test to determine if a service, host, or device is contactable over the network. When creating a TCP Port-based synthetic transaction, you can determine the following:

  • Can the target host be contacted?
  • Is the connection to the target host accepted?
  • Has the connection to the target host timed out?
  • Can the FQDN of the target host be resolved?

When configuring the TCP Port-based synthetic transaction, you specify the computer name or IP address, and the port that you want the transaction to connect to. Figure 7 shows a synthetic transaction that will connect to host Smtp.tailspintoys.internal on port 25.

clip_image008

FIGURE 7 Target and port

More Info: TCP Ports

You can learn more about performing tests against TCP Ports at http://technet.microsoft.com/en-us/library/hh457544.aspx.

UNIX or Linux log file

The UNIX or Linux log file synthetic transaction type allows you to check whether specific text is detected in a certain log file residing on a computer running the UNIX or Linux operating systems. When configuring the synthetic transaction, you must provide the following information:

  • Computer Name This is the name of a computer running UNIX or Linux that has the Operations Manager agent installed and which hosts the log file that you want to monitor. As an alternative, you can specify a Computer Group, which will allow you to use the synthetic transaction to monitor multiple computers with the UNIX or Linux operating system.
  • Log file path The path to, and name of, the log file
  • Regular expression This is a regular expression to detect the text that must occur in the log file to trigger an alert. If the text is a simple string, you don’t have to use a regular expression.

More Info: UNIX or Linux Log Files

You can learn more about performing tests on UNIX or Linux log files at http://technet.microsoft.com/en-us/library/hh457589.aspx.

UNIX or Linux Process

You can use the UNIX or Linux Process synthetic transaction type to determine if a particular process is currently running on a computer running the UNIX or Linux operating systems that have the Operations Manager agent installed. When configuring a synthetic transaction type to determine if a process is running on a computer with a supported UNIX or Linux operating system installed, you must provide the following information:

  • Process name The name of the process.
  • Computer group The Operations Manager computer group that contains the UNIX or Linux hosts that you want to check for the process.
  • Alert sensitivity The sensitivity of the alert to raise if the process is not running.

You can also configure a regular expression to filter process arguments to separate multiple instances of a process with the same name.

More Info: UNIX or Linux Processes

You can learn more about performing tests against UNIX or Linux Processes at http://technet.microsoft.com/en-us/library/hh457572.aspx.

Web application availability

A web application availability synthetic transaction allows you to create a monitoring test for one or more web application URLs to determine that they respond to basic requests. To create a web application availability synthetic transaction, perform the following steps:

1. Click the Management Pack Templates node in the Authoring workspace of the Operations Manager console, and then click Add Monitoring Wizard on the ribbon.

2. On the Monitoring Type page, select Web Application Availability Monitoring, as shown in Figure 8, and click Next.

clip_image009

FIGURE 8 Web application availability monitoring

3. On the General page of the Add Monitoring Wizard, provide a name for the synthetic transaction monitor, and specify an unsealed management pack in which to store the transaction’s settings.

4. On the Enter URLs To Be Monitored page, specify the website name and the website address. Figure 9 shows the default IIS site of the server named Orchestrator.tailspintoys.internal.

clip_image010

FIGURE 9 URLs to monitor

5. On the Where To Monitor From page, specify which computers that have the Operations Manager agent installed will function as watcher hosts.

6. On the View And Validate Tests page, click Run Test to verify that the synthetic transaction works, as shown in Figure 10.

clip_image011

FIGURE 10 Test results

7. Complete the wizard to create the synthetic transaction.

More Info: Web Application Availability

You can learn more about performing tests against web application availability at http://technet.microsoft.com/en-us/library/hh881883.aspx.

Web application transaction

A web application transaction synthetic monitor goes further than a web application availability synthetic transaction as it not only verifies that the target web application is available, but that the web application responds to specific prompts and inputs, including authentication.

You can use the Web Recorder to record a browser session that includes multiple requests to a target web application. You can then use the information generated by the Web Recorder as the transaction used in the synthetic monitor.

To record a web application session and then create a synthetic monitor based on that information, perform the following steps:

1. In the Authoring workspace of the Operations Manager console, select Web Application Transaction Monitoring under Management Pack Templates.

2. In the Tasks menu, click Custom Actions, and then click Record A Browser Session.

3. On the Web Application Editor, shown in Figure 11, provide a name and select an unsealed management pack in which to store the session information.

clip_image012

FIGURE 11 Web application editor

4. On the Web Application Editor - Browsing Session dialog box, click Start Capture. Internet Explorer will launch. Browse to the web application and perform the interaction that you want to test.

5. When you complete the browsing session, close the browser, click Stop Capture, click Apply on the Web Application Editor, then select a watcher node, and click Apply again. Verify that the Web Application Data Imported Successfully message is displayed, as shown in Figure 12, and then close the Web Application Editor dialog box.

clip_image013

FIGURE 12 Web application editor

Clicking Apply creates the web application synthetic transaction. You can also create the web application synthetic transaction by performing the steps in the wizard.

More Info: Web Application Transactions

You can learn more about performing tests against web application Transactions athttp://technet.microsoft.com/en-us/library/hh457553.aspx.

Windows Service

You can create a Windows Service synthetic transaction to determine the state of a service running on a Windows-based computer. When configuring the transaction, you need to provide the following information:

  • Service name
  • Operations Manager computer group

Once the service is selected, you can configure alerts to be triggered if specific CPU and memory thresholds are exceeded.Figure 13 shows a Windows Service synthetic transaction where an alert will be triggered if the CPU utilization exceeds 80 percent and the memory utilization exceeds 150 MB.

clip_image014

FIGURE 13 Performance data

More Info: Windows Service

You can learn more about performing tests against Windows Services at http://technet.microsoft.com/en-us/library/hh457595.aspx.