Starting a database trace

The Database Trace tool produces a text file containing the SQL statements, rule cache hit statistics, timings, and other data that reflect your requestor session's interactions with the Pega Platform database or other relational databases. You can select the database options that are traced. Familiarity with SQL is required to interpret the output.

1. In the header of Dev Studio, click Configure > System > Performance > Database Trace.
2. Click Trace Options.
   The following trace options are available:

   Label	Operation
   Add Batch	Causes the output to include a line for each add batch operation.
   Prepared Statement Query	Causes the output to include a line for each query (SQL database read) operation.
   Prepared Statement Update	Causes the output to include a line for each update (SQL database insert, update, or delete) operation.
   Prepared Statement	Each read, update, or delete operation requested by the Pega Platform server results in a prepared statement query sent to the Pega Platform database. Causes the DB Trace output to include a line for all three types of prepared statements.
   Get Database Metadata	Causes the Database Trace output to include a line for each check of the rule cache. Checks typically occur first time that a table is accessed after startup and before read operations.
   Read blob	Causes the DB Trace output to include a line each time the Storage Stream column ( pzPVStream ) in a row is read. The size of the entry in the column is also reported.
   Write blob	Causes the Database Trace output to include a line each time Storage Stream column ( pzPVStream ) in a row is written.
   Commit	Causes the Database Trace output to include a line each time an internal database commit operation completes. (This is not directly related to the Pega Platform Commit method.)
   Rollback	Causes the Database Trace output to include a line when a PRPC Commit operation is in process but a database rollback operation occurred.
   Assign Connection to Thread	Causes the Database Trace output to include a line when a database connection in a pool is reassigned to a new requestor thread.
   Cache Hit (found)	Causes the Database Trace output to include a line when a rule cache hit occurred and the requested row was found.
   Cache Hit (not found)	Causes the Database Trace output to include a line when a rule cache hit occurred and the requested row did not exist. (For rules, the database cache records both the requests not found and successful requests.)
   Cache Miss	Causes the Database Trace output to include a line when a rule cache miss occurred.
   Run List	Causes the Database Trace output to include a line when a list rule (Rule-Obj-List rule type) was processed by the Obj-List method in an activity.
   List Result Count	Causes the Database Trace output to include a line when a list rule completed and indicates the number of rows returned.
   Activity Start	Causes the Database Trace output to include a line when an activity execution starts.
   Activity End	Causes the Database Trace output to include a line when an activity execution completes.
   Assembly Start	Causes the Database Trace output to include a line when rule assembly starts.
   Assembly End	Causes the Database Trace output to include a line when rule assembly end.

3. Click Trace to turn on tracing for the selected operation.
   Select only the minimal set of options that are required to aid your debugging.
4. Optional: Click Generate stack trace to generate a full Java stack trace.
5. Click Submit.
   The system saves this information in the pyDBTraceSettings page within your requestor page.
6. To start tracing, on the Database Trace tab, click the Click to Start icon (green arrow).