While we are in the process of building RRAS up for our environment (and the ones working on that bit of the environment have been referred by me to this thread), I ran into this issue on our 2019 build 1809 server running our backup solution. Getting dropped a load of notifications on backup failures out of the blue (mostly centered on connection errors).
The application starts and is accessable as a web-page from my client system normally. Getting an RDP session to the server also isn't an issue... But as soon as I start an actual backup it apparently does something in the background.
The webpage drops dead with the notification : Site unreachable: ERR_CONNECTION_RESET
The RDP at that point is also dropped
After about 2 minutes a refresh of the web-page and a re-establishment of the RDP session seems to work normally again, and the backup at that point seems to be active. And the only thing done yesterday was installing this specific fix (KB5014692) yesterday on both servers.
As an aside, as part of the disk-verification mechanism used by the application both HyperV and RRAS services are installed and active on both servers. So that does tie in with the reports on RRAS and possibly HyperV.
I've deinstalled this update and that does seem to sort the problem. So far I have not run into problems on any of our other servers (RRAS or otherwise, tho I did emphasize this issue seems to only hit RRAS installed servers), but I've given notice to most everyone involved that this patch may be causing issues, so they're not stuck wondering how and what but can atleast try if a deinstallation sorts the issue.
If it does I hope Microsoft will get on this, and sort this issue for the next CU in July.