SETTING THE SYSTEM CONFIGURATIONXEROX DOCUPRINT 4850/4890 IPS GUIDE TO CONFIGURING AND MANAGING THE SYSTEM 4-255. Enter a name for the trace file. You can enter the full directorypath along with the file name to save the trace file directly to adesired location. If no file name or file path is specified, the fileis saved as Trace and stored in the /var/db directory.Note: It is recommended that the trace file name start withtrace_, followed by the name of the trace, all in lower casecharacters. For example, if the trace name is test, enter thefollowing as the file name:trace_testNote: The trace file is a large file and can quickly fill up adirectory. For that reason, the default /var/db directory is therecommended place to store trace files.6. Click OK to start the trace.7. Start the printer at the host.8. Send a job to the IPS. The trace runs on this job. (Thethroughput speed of the job when running in Trace Mode isslower than in normal operation.)9. Select Normal Operation from the Trace Options pull-downmenu.10. Click Quit.Executing a captured traceTo execute a trace to assist with system diagnostics:1. Capture the trace, following the steps described in the section“Capturing a trace” above.2. Select Trace Options from the Diagnostics pull-down menu onthe main window.3. Click Offline to disable the interface.4. From the Trace Mode pull-down menu, select Execute Trace.The Default Trace File Name prompt window displays.5. If you have already given the trace file a name during thecapturing process, that name is displayed on this screen.If you were not capturing a trace file, or if you want to execute atrace file different from the one just captured, enter the name onthe line provided in the Default Trace File Name window.6. Click Apply. The job prints. After the job has completed, thesystem returns to normal operation.7. After the job has completed, select Normal Operation.8. Click Online to enable the interface.9. Click Quit.Note: After executing a trace file, make sure to cancel the FSS andrestart IPS before submitting any jobs. Otherwise, the output of anysubmitted jobs may be missing all of the text data.
