The actual full error on my system is:

2015/06/03 12:47:17 [2012] rsync: read error: Connection reset by peer (104)
2015/06/03 12:47:17 [2012] rsync error: error in socket IO (code 10) at io.c(785) [sender=3.1.1]
2015/06/03 12:57:18 [316] sent 0 bytes  received 0 bytes  total size 0

Inside the BackupPC XFER Error Log:

Sending args: --server --sender --numeric-ids --perms --owner --group -D --links --hard-links --times --block-size=2048 --recursive . .
Xfer PIDs are now 25110
[ skipped 5027 lines ]
Read EOF: Connection reset by peer
Tried again: got 0 bytes
Can't write 4 bytes to socket
finish: removing in-process file path/sanitized/for/your/viewing/pleasure/UVHD-S-1-5-21-773923596-1703164468-81712224-1105.vhdx
[ skipped 7 lines ]
Child is aborting
Done: 231 files, 38884879 bytes
Got fatal error during xfer (aborted by signal=PIPE)
Backup aborted by user signal

You'll notice this is a VHDX file - 40GB in size this one is. The problem is indeed a timeout, but it's not at the BackupPC end, it's at the rsyncd end. I added the following line to the proper module of my rsyncd.conf on the host being backed up:

 timeout = 10800

Restart rsyncd on that box and now it works like a champ.