The following section provides information on the different approaches using Visual Studio that depend on what the application
to debug is:
- Attach to a running process - use Just-In-Time (JIT) debugging to enable the debugger to connect to a running application
when an error occurs.
- Create and debug a core dump file - when an application crashes, you can arrange for its state to be saved to a core dump
file, which can indicate where the error occurred in the source code, the contents of memory at the time of the error, and
the values of any variables and expressions set at the time.
- Debug link libraries - either directly, if the project only compiles to a single .dll file or, for a project that compiles
to multiple .dll files, by specifying one of the .dll files as the
Start external program option.
- Attaching to a running program - you use the Wait for a debuggable attachment option. With it the debugger waits for the program
to start in a separate process before attempting to attach to it.
- Catching memory overflows - you can set COBOL watchpoints on data items to watch the area of memory associated with these
data items.
- Debugging applications running on a remote machine.