AnsweredAssumed Answered

Fixing UFL / Backfilling from CSV

Question asked by cstation on Mar 3, 2017
Latest reply on Mar 6, 2017 by cstation

My question is:  What would be the best way (or any way) to get UFL to reprocess my CSV data files to fill in the missing data.  I have tried just running the UFL interface again, with a corrected INI and updated csvfiles, but it doesn't appear to add any more data into the archive. 




I was attempting to backfill roughly 7 days of data, sampled at 5 seconds for 173 data points.   It has been several years since I had to backfill a PI server, and have spent most of the week watching youtube vidoes, reviewing white papers and reading manuals.  Of course, I messed up the time format in the UFL INI file, and didn't notice until I had completed all 173 tags (I only checked the first 100 or so values after my test import).  The problem was (I am assuming) with my date time format.  It only imported data that had a timestamp that started with 10/11/12 (2 digits).  I went ahead and reprocessed the CSV files in C# using ToString("MMM/dd/yyyy HH:mm:ss") to ensure the datetime was consistent.  I spot checked a few files, and the time stamp appears to be correct (making sure that 11PM showed up as 23:00:00, 1AM showed up as 01:00:00). 


So, I figured, just rerun the UFL again with the updated INI and data files, and everything would be good to go.  It wasn't, and I am sure I am missing something simple.  There should be 140k points for each tag, but the archive editor only shows around 25k for the tags I reviewed.  I have tried changing the Archive mode from the UFL Configuration Utility (ArcReplace), but still do not get any additional data.


I reviewed any notes I could find online and the forums, but couldn't find anything that helped, as I am probably missing the right keyword in my search.  If I have to delete the archives and start over again, that is fine, if you could recommend a white paper on the best method to do so, that would be great.. 


Here are some screen shots of the data, and my INI file.  Also, a screen shot of the data exported from the archive (via datalink) compared to the raw CSV file.