This article from InSource shows how the Historian in earlier versions may try to backfill your History Blocks from an earlier date to the first known History Block. This article will show how to resolve this issue.
- Author: Glenn Yancey
- Published: 06/21/2017
- Applies to: Historian Server 2014 and earlier
You can insert and update data by importing it into the history blocks, as long as the data is formatted according to a specific comma-separated values (CSV) format.
When doing a CSV File import, you need to make sure that the header row has the correct parameters or you might start to see thousands of folders getting created in your CIRCULAR directory. That might start causing your Historian to behave slower than normal at times.
In this example, I show how I have taken data from 1985 and then imported that information via a CSV file into the Wonderware Historian. If i just left the headers as they were since I copied from a known example, the Historian would create a History block for the data inserted in October 15, 1985, and also recreate History Blocks all the way to the very first History Block stored by the Wonderware Historian. Again, resulting in THOUSANDS of History Block folder filling up your circular directory.
This can be resolved by changing the parameter that represents the "Time Span of Replacement Blocks". In Wonderware Historian 2014 and earlier, you can change it to "2" and it will not create replacement History blocks between the earlier date for which the data was inserted up to the first stored History Block. Starting with Wonderware Historian 2014 R2, the Time Span of Replacement Blocks and Block Behavior fields are no longer applicable. You must provide a value for these fields, but they are ignored. So in other words with 2014 R2, you will not have constant History Blocks being backfilled or replaced between the earlier date for which the data was inserted up to the first stored History Block.