Changing the clock on the computer probably caused the history system to think the system was down for some period of time, in which case the values returned will be of “bad” quality, and won’t display on the chart. If the gap is small enough, the chart still draws lines, and could possible look like what you have.
Take a look at the “sqlth_sce” table- it holds the record of when the scan classes were running, and is used to calculate when the system was down. While the system is up, each scan class updates the “end_time” of the most recent entry for it’s id. The thing to determine is whether there is a gap in this table represented by multiple rows for a given “scid”, with a difference between the end_time of one and the start_time of the next, that covers the time you’re having a problem with.
Of course, working with these tables by hand can be a tad difficult due to the way timestamps are stored. If you’re using mysql, here’s a query to view the entries in order with easy to read dates:
SELECT scid, from_unixtime(start_time/1000) as 'start', from_unixtime(end_time/1000) as 'end' FROM sqlth_sce
If you see a gap that corresponds to your point, the easiest thing to do would be to make the closest entry cover it. For example, update the “start_time” of the last row to contain it. It’s in milliseconds, so if you wanted to expand it by a day, you could do something like:
UPDATE sqlth_sce SET start_time=start_time-(60*60*24*1000) WHERE start_time=X and scid=Y
With the appropriate values for scid and the previous start time value.
Hope this helps,