It would appear that the default behavior is to suppress all historical interface data when the interface is downed on a given device.
This is causing issues with our operations teams, we need to address this as soon as possible.
The default component filter on the Network Interface Monitoring Profile is set to:
Administrative Status does not equal down(2)
By default we do not synch filtered interfaces to CAPC or odata.
Create your own Monitoring Profile with own filter that doesn’t filter down interfaces.
So you can continue polling them even when down. And they will remain synced with PC unless it fails whatever new filters you apply.
I would use the MP with care, applying only to mission critical devices and carefully monitoring how many i\f’s are added and what impact it has.