Hacker News Reader

Comment by djhaskin987 | Hacker News Reader

I've only ever heard of mongo successfully used for two use cases:

1. As a cache, like redis 2. As a log store, like Elasticsearch

In both cases, the data is somewhat ephemeral, and not the "source of truth" for the app. The minute it is used for holding real, customer supporting data, things start to get dire real fast.

View on HN

Plenty of people are successfully using MongoDB for real, customer supporting data at a large scale. There's a selection of users on the website for a start: https://www.mongodb.com/who-uses-mongodb

(Disclaimer: I work for MongoDB)

But we'll never know if those companies will be better off if they hadn't. You're just making a bandwagon argument

(Disclaimer: I hate MongoDb)

I have used Mongo to store medical data that had to be stored for 7 years. We set it to make sure that all three servers in the cluster did a write acknowledgement.

Thank you! As an aside, is there a reason you're not using a write concern of "majority"? If you have w:3 in a three node cluster then if one node goes down, writes are going to start throwing wtimeout errors (assuming you have wtimeout set) even though the data may safely be written to a majority of nodes. We generally recommend setting w:"majority" for this reason.

It was recommended by our vendor. I asked why, he was just paranoid. But, most of the data was entered on a semi connected mobile device, stored on the handheld and synced when they had service. If it wasn’t immediately writable to the server, it wasn’t a show stopper.

Reply on news.ycombinator.com