actually I watched the video and the errors happend here are more cultural errors. I guess the switch is more like a "we failed, so we start new". and new means with something different.

also their replication strategy looked like a joke and not enough automation. hopefully they never use galera, else their bad engeering practices could actually suffer in huge data losses.

I've once run galera-cluster and it pretty easy came to data losses especially after short network splits which occured randomly at the network.