How to avoid data loss on server failure with MongoDB on a single machine?

Question

I have read that mongoDB don't write data to disk right away, it does this periodically.

Any thoughts on how to deal with this?

Answer 1

You can enable journaling with --journal.

Check out http://www.adathedev.co.uk/2011/03/mongodb-journaling-performance-single.html and http://www.mongodb.org/display/DOCS/Durability+and+Repair

Answer 2

Besides --journal that is enabled by default since MongoDB 2.0 (only on 64 bit machines), there is a flag that you can set when persisting data:

• safe => false: do not wait for a db response
• safe => true: wait for a db response
• safe => num: wait for that many servers to have the write before returning
• fsync => true: fsync the write to disk before returning. fsync => true implies safe=>true, but not visa versa.

If fsync=>false and safe=>true and the write could be in successfully applied to a mmapped file but not yet written to disk