This is free sandbox mode. The db reads and writes are never overloaded. Its been 6 hours and its still down.
Without at least 2 replicas, no primary will be set, and thus I cannot use the cli or anything else. The entire cluster is in unavailable mode! Its been 6 hours! Imagine if this was in prod!
Idk what to do. atlas tries to restart it, then a primary gets set, but then a replica tries a rollback, causing only 1 replica left, and this loop repeats! I can’t shut down the entire thing because since only 1 replica is available I can’t execute any commands and the only option on atlas is a big button telling me to “UPGRADE”.