I have a RRDTool database that has data inside and I want to be able to import this data into another RRDTool database that differs only by the RRAs.
I want to increase the precision of historical data, so I tried how to grow the RRA via rrdresize but it doesn't recompute the added rows. That means I get strange results when graphing as various "zoom levels" : the graph has data when graphing at a lower precision, but empty when I increase the precision since RRDTool automatically selects the best precision level, and the newly added RRA rows are empty (UNKNOWN).
I cannot use rrdrestore either since that also creates the RRD (and doesn't just fill an already existing one).
I know that I cannot recreate the lost data, but I would like to at least copy the data from the lower precision RRA to the new high precision one.
I did write a simple tool, rrdmove, that just loops on every step to do a rrdfetch/rrdupdate cycle so that RRDTool just recomputes all the RRA in the new database, but it does not seem very efficient.