Killing Long Running Ops More Safely in MongoDB

I recently read a great summary of the upcoming features in MongoDB 2.6 by Simon Maynard from bugsnag, but was a bit worried by the sample code he shared that he uses to kill long running operations. On a replica set, it stands a chance of breaking replication, and potentially other things.

Hence, I put a safer version up as a gist - it's a small improvement, but potentially a crucial one if you are going to be running it on a production system.