Working on more reliable dynamo to pg transfer scripts that target devices directly using batch inserts and take serial number and date ranges as input. No more reliance on finding missing values. Just clear hole and fill it approach with good logging for serials and timestamps.
This effort has been placed on the back burner because the original motivation was based on a believe the analyze-sync, transfer-data scripts in place were badly broken. Now however it appears this was based on incorrect interpretation of data. For example Lansing group has stopped on 3/1. There were also other confusing items.
Working on more reliable dynamo to pg transfer scripts that target devices directly using batch inserts and take serial number and date ranges as input. No more reliance on finding missing values. Just clear hole and fill it approach with good logging for serials and timestamps.
This effort has been placed on the back burner because the original motivation was based on a believe the analyze-sync, transfer-data scripts in place were badly broken. Now however it appears this was based on incorrect interpretation of data. For example Lansing group has stopped on 3/1. There were also other confusing items.