Hi tgif,
Welcome to the wonderful world of log parsing.
LV doesn't like to graph categorical data (event 1, event 2) as well as it likes to graph continuous data (1, 2, 3... ). This means that one possible solution is to convert your categorical data to numeric data. The problem is, for that to work, you'll need to know at least most of your events in advance.
Be that as it may, there are some things that will make your life easier: your date/time stamp in your log should always be in the same place/format. Your events should be easy to recognize (delimited with tabs or other unique characters, for instance). Data corruption should be minimal (some log writers have real problems with that).
In any event, my version of sample code is attached.
Good luck (and <wave> to Dennis),
Joe Z.