The SEWERSUBNETLINE feature class is growing quite large within our eGDB. The branch-versioned feature service requires archiving to be turned on, so every time we update our "dirty" subnetworks, the previous subnetworks are kept indefinitely. The geoprocessing tools to remove archive records won't work on an eGDB feature class.
The SEWERSUBNETLINE currently holds 72 distinct subnetworks (in some cases with dozens of archived records). The table alone is now > 1.5GB in size.
Can we get some kind of tool or script that can "clean out" the archived records periodically (or even selectively)?