"xxxx Full Hash incorrect" on fresh sync to testnet

[WhoSoup]

This occured on a node testing the new p2p package on the testnet during a fresh sync but it seems to be an issue that's been around for a while:

xxxx Full Hash incorrect at height 78
panic: Full hash incorrect block at 78 FNode0

goroutine 90 [running]:
github.com/FactomProject/factomd/state.(*State).LoadDBState(0xc4200fa000, 0xc40000004f, 0x2, 0x2, 0x0, 0x0)
    /home/ubuntu/go/src/github.com/FactomProject/factomd/state/state.go:1284 +0x17f1
github.com/FactomProject/factomd/state.LoadDatabase(0xc4200fa000)
    /home/ubuntu/go/src/github.com/FactomProject/factomd/state/loadDatabase.go:76 +0x2f0
created by github.com/FactomProject/factomd/engine.startServers
    /home/ubuntu/go/src/github.com/FactomProject/factomd/engine/NetStart.go:619 +0x15f

The error is triggered when loading directory blocks and the "getprevkeymr" doesn't match up with the previously loaded block's full hash. There are numerous reasons why this could fail and my first inclination would be to say it's a data transmission (either from the network or the disk) error, such as a flipped bit.

I have found several previous posts about this error in discord from previous versions of the software:
6.1.1 https://i.imgur.com/gGQEkQH.png
5.x.x https://i.imgur.com/7P6vzIP.png
bandaid https://i.imgur.com/Zrf7Co5.png

It seems to be happening on TestNet only and very sporadically. Has anyone ever encountered this error on mainnet? Is there any other information available on this?