11.0 Advanced Topics
This section covers some of the more advanced features of the Upgrade Assistant. Scheduling is not used extensively but it is offered for the use cases that require it. There are also some advanced settings that may be necessary for certain use cases.
11.1 Scheduling Data Gathering
In most cases you will want to monitor the execution of the steps to produce and compare the trace files, but if you want to run the steps off-hours or on the weekend it might be useful to schedule the execution of the capture and compare steps. The Schedule Settings screen can be used to schedule execution.
In almost all cases you will want to schedule the capture and compare process to execute once, but if you want to schedule it to run periodically you can set this to repeat at a specified interval. The Enabled checkbox allows you to enable or disable the scheduled execution.
Set the date and time the job should begin execution.
The Trace Duration control sets the length of time the replay capture trace should run. The default is one minute which is probably too short to gather a representative trace. The Job Step Interval configures the delay between job steps. The one minute default is normally adequate. The Period Time setting isn’t relevant for Upgrade Assistant scheduling.
11.2 Advanced Settings
With the exception of the Maximum Trace File Size, the defaults of these settings will work in most cases.
This sets the maximum size of a trace file. Very large trace files can cause very long execution times and timeout errors.
This sets the number of threads to use for replaying the capture file. In most cases this should be left at the default of 1 but if you have a large trace file you may need to increase this. The order of replayed events is not necessarily maintained with multiple replay threads so you may need to use a single replay thread to avoid differences caused by ordering between threads.
The replay engine has a health monitor that monitors execution time for processes and kills processes that have been running longer than the configured Wait Interval time. A setting of 0 means let all processes run as long as necessary.
This value configures how often the health monitor will check for processes running longer than the Wait Interval.
This sets the max time to wait for a query to complete. The default of 0 means wait forever and should work in almost all cases.
The compare function uses temporary files to hold the difference files. This sets the location where these files are stored. Once you have identified the differences between the versions, you can go here for help in fixing the issues: 12.0 Fixing Differences If you have issues with running or interpreting the results of you can get help here: 13.0 Support