Every database is an operational nightmare. Just wait until MySQL segfaults every time it gets more than 4 SSL connections at once, or takes four hours to warm up its inno cache. Or Cassandra compaction locks up a node so hard its peers think its down, turning your cluster into a recursive tire fire of GC. Or a Riak node stops merging bitcask and consumes all your disk in a matter of hours. Or Mongo ...
My point is that every database comes with terrifying failure modes. There is no magic bullet yet. You have to pick the system which fits your data and application.
Magic bullet? These systems are the closest you can find in the world of software.
[Edit:] Don't get me wrong, Oracle's DBs are serious business; an incredible feat of engineering. If I could afford them I'd probably use them more often. But everything in this business is a cost tradeoff--in licenses, in dev time, in risk.
How often did you manage to spectacularly fail PostgreSQL to qualify it an 'operational nightmare'?