Some limitations and bugs in MongoDB, mostly related to its MapReduce and import/export:
There’s still plenty to like in Mongo, but at this point, I feel like Mongo’s mapReduce functionality is better suited to running queries which are too big to fit in memory, rather than serious data crunching. Perhaps my difficulties have been due to getting sharding involved with mapReduce. It’s also possible I’ve made a crucial mistake in configuring sharding, but I think I followed the directions pretty closely.
Original title and link: Limitations of MongoDB (NoSQL databases © myNoSQL)