So.....
Cassandra vm had a crash yesterday...
2 nodes with rep factor 1. (FML)

One node wouldnt start... Eventually found out one of the commit replays had an exception (the one at the time of the crash).

Boss trying to push me towards a fix all this time which was:
"Let's delete the vm and have Cassandra running on one vm"

There are not we enough curses in the world.
🖕🖕🖕🖕🖕🖕🖕🖕🖕🖕🖕

P.S. there are no backups.