SQL Server

SQL Server 2008 : SQL Server Profiler - Creating Traces

2/28/2011 9:23:16 AM
Because SQL Server Profiler can trace numerous events, it is easy to get lost when reading the trace output. You need to roughly determine the information you require and how you want the information grouped. For example, if you want to see the SQL statements that each user is submitting through an application, you could trace incoming SQL statements and group them by user and by application.

When you have an idea about what you want to trace, you should launch the SQL Server Profiler by selecting Start, then SQL Server 2008, then Performance Tools, and finally SQL Server Profiler. You also can launch it from within SSMS from the Tools menu. When you launch the Profiler, you are presented with an application window that is basically empty. To start a new trace, you select the File menu and choose New Trace. In the connection dialog that is displayed, you can enter the connectivity information for the server you want to trace. After the connection is established, the General tab of the Trace Properties window (see Figure 1) is displayed.

Figure 1. General trace properties.

The first place you should look when creating a new trace is at the trace templates. These templates contain predefined trace settings that address some common auditing needs. They have preset events, data columns, and filters targeted at specific profiling scenarios. The available trace templates, found in the template drop-down on the General tab of the Trace Properties window, are listed in Table 1.

Table 1. SQL Profiler Templates
SP_CountsTracks all the stored procedures as they start. No event except for the stored procedure starting is traced.
StandardTraces the completion of SQL statements and Remote Procedure Calls (RPCs) as well as key connection information.
TSQLTraces the start of SQL statements and RPCs. This template is useful for debugging client applications where some of the statements are not completing successfully.
TSQL_DurationTraces the total execution time for each completed SQL statement or RPC.
TSQL_GroupedTraces the start of SQL statements and RPCs, grouped by Application, NTUser, LoginName, and ClientProcessId.
TSQL_LocksTraces the completion of SQL statements along with the key lock information that can be used to troubleshoot lock timeouts, deadlocks, and lock escalation issues.
TSQL_ReplayCaptures profiling information that is useful for replay. This template contains the same type of information as the standard template, but it adds more detail, including cursor and RPC output details.
TSQL_SPsTraces stored procedures in detail, including the start and completion of each stored procedure. The SQL statements within each procedure are traced as well.
TuningPerforms a streamlined trace that tracks only the completion of SQL statements and RPCs. The completion events provide duration details that can be useful for performance tuning.

Keep in mind that the templates that come with SQL Server 2008 are not actual traces. They simply provide a foundation for you in creating your own traces. After you select a template, you can modify the trace setting and customize it for your own needs. You can then save the modified template as its own template file that will appear in the template drop-down list for future trace creation.

Trace Name is another property you can modify on the General tab. Trace Name is a relatively unimportant trace property for future traces. When you create a new trace, you can specify a name for the trace; however, this trace name will not be used again. For instance, if you have a trace definition you like, you can save the trace definition as a template file. If you want to run the trace again in the future, you can create a new trace and select the template file that you saved. You will not be selecting the trace to run based on the trace name you entered originally. Trace Name is useful only if you are running multiple traces simultaneously and need to distinguish between them more easily.


Do yourself a favor and save your favorite trace definitions in your own template. The default set of templates that come with SQL Server are good, but you will most likely want to change the position of a column or add an event that you find yourself using all the time. It is not hard to adjust one of the default templates to your needs each time, but if you save your own template with exactly what you need, it makes the task all the more easy. After you save your own template, you can set it as the default template, and it will be used by default every time you start the Profiler.

The Save to File and Save to Table options on the General tab of the Trace Properties page allow you to define where the trace output is stored. You can save the output to a flat file or SQL Server table.

The last option on the General tab of the Trace Properties window is the Enable Trace Stop Time option. This scheduling-oriented feature allows you to specify a date and time at which you want to stop tracing. This capability is handy if you want to start a trace in the evening before you go home. You can set the stop time so that the trace will run for a few hours but won’t affect any nightly processing that might occur later in the evening.


The events and data columns that will be captured by your Profiler trace are defined on the Events Selection tab. An example of the Events Selection tab is shown in Figure 2.

Figure 2. The Events Selection tab.

The Events Selection tab consolidates the selection of events, data columns, and filters on one screen. One of the biggest advantages of the SQL Server 2008 Events Selection tab is that you can easily determine which data columns will be populated for each event by looking at the columns that have check boxes available for the event. For example, the Audit Login event has check boxes for Text Data, ApplicationName, and others but does not have a check box available for CPU, Reads, Writes, and other data columns that are not relevant to the event. For those data columns that have check boxes, you have the option of unchecking the box so that the data column will not be populated for the event when the trace is run.

You may find that adding events in SQL Server 2008 is a bit confusing. When you select a template, the event categories, selected events in those categories, and selected columns are displayed in the Events Selection tab. Now, if you want to add additional events, how do you do it? The answer to this question lies in the Show All Events check box in the lower-right corner of the Events Selection tab. When you click this check box, all the available event categories are listed on the screen. The events and columns that you had previously selected may or may not be visible on the screen. They are not lost, but you may need to scroll down the Events Selection tab to find the event categories that contain the events you had selected prior to selecting the Show All Events check box.

You will also notice that all the events in the categories in which you had events selected are displayed. In other words, if you had only 2 events selected in the Security Audit category and then selected the Show All Events check box, you see all 42 events listed. The only 2 events selected are the ones you had selected previously, but you need to wade through many events to see them. One upside to this kind of display is that you can easily view all the events for a category and the columns that relate to the events. One possible downside is that the Events Selection tab can be very busy, and it may take a little extra time to find what you are looking for.


If you capture too many events in one trace, the trace becomes difficult to review. Instead, you can create several traces, one for each type of information that you want to examine, and run them simultaneously. You can also choose to add or remove events after the trace has started. Keep in mind that you can pause a running trace, change the selected events, and restart the trace without losing the output that was there prior to pausing the trace.

Your ability to select and view events is made easier by using the tree control available on each event. The tree control allows you to expand or compress an event category. When you click the + icon next to a category, all the events are displayed. When you click the – icon, the event category is collapsed to a single row on the display. When an event has been selected for use within a category, the category name is shown in bold. If you want to add all the events in a category to your trace, you can simply right-click the category name and choose the Select Event Category option. You can also remove all events in a category by right-clicking the category name and choosing the Deselect Event Category option.

Understanding what each of the events captures can be a challenging task. You can refer to “SQL Server Event Class Reference” in Books Online for a detailed description, or you can use the simple Help facility available on the Events Selection tab. The Events Selection tab has a Help facility that describes each of the events and categories. The Help text is displayed on the Events Selection tab below the list of available events. When you mouse over a particular event or event category, a description of that item is shown. This puts the information you need at your fingertips.


If you are going to use SQL Server Profiler, you should spend some time getting to know the events first and the type of output that Profiler generates. You should do this first in a development environment or standalone environment where the Profiler’s effect on performance does not matter. It’s a good idea to start a trace with a few events at a time and execute some relevant statements to see what is displayed for each event. You will soon realize the strength of the SQL Server Profiler and the type of valuable information it can return.

Data Columns

The columns of information captured in a Profiler trace are determined by the Data Columns selected. The Events Selection tab has the functionality you need to add columns, organize the columns, and apply filters on the data returned in these columns. As mentioned earlier, you can select and deselect the available columns for a particular event by using the check boxes displayed for the listed events. To understand what kind of information a column is going to return, you can simply mouse over the column, and Help for that item is displayed in the second Help box below the event list. Figure 3 shows an example of the Help output. In this particular case, the mouse pointer is over the ApplicationName column returned for the SQL:BatchCompleted event. The first Help box displays information about the SQL:BatchCompleted event, and the second Help box shows information about the data column.

Figure 3. Help for data columns on the Events Selection tab.

Keep in mind that there is a default set of columns displayed for each event. You can view additional columns by selecting the Show All Columns check box. When you choose this option, an additional set of columns is displayed in the Events Selection tab. The additional columns are shown with a dark gray background, and you may need to scroll to the right on the Events Selection tab to be able to see them. Figure 4 shows an example of the additional columns displayed for the Performance event when the Show All Columns option is used. Some of the additional columns available for selection in this example are BigintData1 and BigintData2.

Figure 4. Additional columns displayed with the Show All Columns option.

To organize the columns you have selected, you can choose the Organize Columns selection on the Events Selection tab. This Organize Columns window allows you to change the order of the columns in the trace output as well as group the data by selected columns. Figure 5 shows an example of the Organize Columns window with the groups and columns selected by default when you use the TSQL_Grouped template.

Figure 5. Organizing columns in the Events Selection tab.

To change the order of a column, you simply select the column in the list and use the Up or Down buttons to move it. The same movement can be done with columns selected for grouping. You add columns to groups by selecting the column in the data list and clicking the Up button until the column is moved out of the Columns list and into the Groups list. For example, in Figure 6.6, you can group the SPID column by selecting it and clicking the Up button until it moves into the Groups tree structure instead of the Columns tree structure.


You can select a particular column for all events by right-clicking the column header in the Events Selection tab and choosing the Select Column option. This causes all the check boxes on the grid to be selected. To remove a column from all events, you right-click the column header and choose Deselect Column.

The number of columns selected for grouping and the order of the columns are both important factors in the way the trace data will be displayed. If you choose only one column for grouping, the trace window displays events grouped by the values in the grouped data column and collapses all events under it. For example, if you group by DatabaseId, the output in the trace window grid displays DatabaseId as the first column, with a + sign next to each DatabaseId that has received events. The number displayed to the right of the event in parentheses shows the number of collapsed events that can be viewed by clicking on the + sign. Figure 6 shows an example of the trace output window that has been grouped by DatabaseId only. The database with a DatabaseId equal to 6 is shown at the bottom of the grid in this example. The grid has been expanded, and some of the 20 events that were captured for this DatabaseId are shown.

Figure 6. Grouping on a single column.

If you select multiple columns for grouping, the output in the trace window is ordered based on the columns in the grouping. The events are not rolled up like a single column, but the trace output grid automatically places the incoming events in the proper order in the output display.


The organization of columns in a trace can happen after a trace has been defined and executed. If you save the trace to a file or table, you can open it later and specify whatever ordering or grouping you want to reorganize the output. This flexibility gives you almost endless possibilities for analyzing the trace data.


Filters restrict the event data returned in your trace output. You can filter the events captured by the SQL Profiler via the Column Filters button on the Events Selection tab. An example of the Edit Filter window is shown in Figure 7. All the available columns for the trace are shown on the left side of the Edit Filter window. Those columns that have filters on them have a filter icon displayed next to the column in the column list.

Figure 7. Editing filter properties.

The filtering options in SQL Server 2008 are similar to those available in SQL Server 2005. Which options are available depends on the type of column you are filtering on. The different filtering options are as follows:

  • Like/Not Like— This option enables you to include or exclude events based on a wildcard. You should use the % character as your wildcard character. When you have completed a filter definition you can press Enter to create an entry space for another filter definition. For example, with the ApplicationName filter, you can specify Like Microsoft%, and you get only those events related to applications that match the wildcard, such as Microsoft SQL Server Management Studio. This filtering option is available for text data columns and data columns that contain name information, such as NTUserName and ApplicationName.

  • Equals/Not Equal To/Greater Than or Equal/Less Than or Equal— Filters with this option have all four of these conditions available. For the Equals and Not Equal To conditions, you can specify a single value or a series of values. For a series of values, you hit enter after each value is entered and a new entry space is created for you to enter the next value. For the other conditional types, a single value is supplied. For example, you can filter on DataBaseID and input numeric values under the Equals To node of the filtering tree. This filtering option is available for numeric data columns such as Duration, IndexId, and ObjectId.

  • Greater Than/Less Than— This type of filtering option is available only on time-based data columns. This includes StartTime and EndTime filters. These filters expect date formats of the form YYYY-MM-DD or YYYY-MM-DD HH:MM:SS.

Each data column can use one of these three filtering options. When you click the data column available for filtering, you see the filtering options for that column displayed in the right pane of the Edit Filter window. You enter the values on which you want to filter in the data entry area on the filter tree. This input area is shown when you select a specific filtering option. For multiple filter values, you press the Enter key after you enter each value. This causes a new data entry area to appear below the value you were on.


Filters applied to columns that are not available or selected for an event do not prevent the event data from being returned. For example, if you place a filter on the ObjectName column and choose the SQL:StmtStarting event as part of your trace, the event data is not filtered because ObjectName is not a valid column for that event. This behavior may seem relatively intuitive, but it is something to consider when you are receiving output from a trace that you believe should have been filtered out.

Also, be careful when specifying multiple filter values and consider the Boolean logic applied to them. When you specify multiple values for the Like filter, the values are evaluated with an OR condition. For example, if you create a filter on ObjectName and have a Like filter with values of A%, B%, and C%, the filter returns object names that start with A or B or C. When you use the Not Like filter, the AND condition is used on multiple values. For example, Not Like filter values for ObjectName of A% and C% result in objects with names that do not start with A and object names that do not start with C.

Other -----------------
- SQL Server 2008 : SQL Server Profiler Architecture
- SQL Server 2008: Administering Database Objects - Working with Tables (part 7) - Partitions
- SQL Server 2008: Administering Database Objects - Working with Tables (part 6) - Compression
- SQL Server 2008: Administering Database Objects - Working with Tables (part 5) - Sparse Columns
- SQL Server 2008: Administering Database Objects - Working with Tables (part 4) - Check Constraints
- SQL Server 2008: Administering Database Objects - Working with Tables (part 3) - Foreign Key Constraints
- SQL Server 2008: Administering Database Objects - Working with Tables (part 2) - Primary Key Constraints & Unique Constraints
- SQL Server 2008: Administering Database Objects - Working with Tables (part 1) - Default Constraints
- SQL Server 2008: Administering Database Objects - Working with Database Snapshots
- Programming with SQL Azure : WCF Data Services (part 3)
- Programming with SQL Azure : WCF Data Services (part 2) - Creating the Client Application
- Using XML in SQL Server 2008: Relational Data As XML - The FOR XML Modes (part 4) - EXPLICIT Mode
- Using XML in SQL Server 2008: Relational Data As XML - The FOR XML Modes (part 3) - AUTO Mode
- Programming with SQL Azure : WCF Data Services (part 1)
- Using XML in SQL Server 2008: Relational Data As XML - The FOR XML Modes (part 2) - Working with Binary Columns
- Using XML in SQL Server 2008: Relational Data As XML - The FOR XML Modes (part 1) - RAW Mode
- Programming with SQL Azure : Connecting to SQL Azure (part 4) - Sqlcmd
- Programming with SQL Azure : Connecting to SQL Azure (part 3) - ODBC
- Programming with SQL Azure : Connecting to SQL Azure (part 2)
- Programming with SQL Azure : Connecting to SQL Azure (part 1) - ADO.NET
- First look: Apple Watch

- 10 Amazing Tools You Should Be Using with Dropbox

- Sigma 24mm f/1.4 DG HSM Art

- Canon EF11-24mm f/4L USM

- Creative Sound Blaster Roar 2

- Alienware 17 - Dell's Alienware laptops

- Smartwatch : Wellograph

- Xiaomi Redmi 2
Popular tags
Microsoft Access Microsoft Excel Microsoft OneNote Microsoft PowerPoint Microsoft Project Microsoft Visio Microsoft Word Active Directory Biztalk Exchange Server Microsoft LynC Server Microsoft Dynamic Sharepoint Sql Server Windows Server 2008 Windows Server 2012 Windows 7 Windows 8 Adobe Indesign Adobe Flash Professional Dreamweaver Adobe Illustrator Adobe After Effects Adobe Photoshop Adobe Fireworks Adobe Flash Catalyst Corel Painter X CorelDRAW X5 CorelDraw 10 QuarkXPress 8 windows Phone 7 windows Phone 8