Ask HN: Can someone explain why this line exists?

25 points by andrewgross 12 years ago | 16 comments

https://github.com/mongodb/mongo-java-driver/blob/master/src/main/com/mongodb/ConnectionStatus.java#L213

Found via: https://twitter.com/avibryant/status/339840598871769088

dampier 12 years ago |

If you read the code, you will notice that the FIRST time this error is encountered, it will always be logged.

The test in question is done frequently, and subsequent failures are about 100 times less interesting than the first. Unless you just like filling up disks with log files while your network is acting up.

Finally, note that the "_ok" flag will be RESET and the resolved condition logged once the test finally returns to succeeding.

andrewgross 12 years ago | |

Awesome, thanks for the explanation

lotyrin 12 years ago | | |

Thank you for an excellent counterexample for when people claim code comments are useless.

zaroth 12 years ago |

I commented earlier, but I can't get over how ridiculous that line is. If you started by just writing the intent (!ok && rand() > .1) you actually have to work really hard to turn it into what you have there, and introduce a bug in the process.

Reminds me of this: http://www.osnews.com/story/19266/WTFs_m - "Dude, WTF!"

byoung2 12 years ago |

If an exception was thrown, and _ok evaluates to false, then approximately 10% of the time, log the exception to the error log. Probably just to cut down on noise in the logs, or to account for brief periods of latency where the server appears down, but isn't.

andrewgross 12 years ago | |

Thanks, was reviewing it with a coworker and crept up on the same conclusion. Seems a bit hard to test and definitely alarming at first glance.

avibryant 12 years ago | | |

I think that you're right about the intent of the author, but unless I'm mistaken, that's not what the code does. If _ok is false, it boils down to if(!(Math.random() > 0.1))) which is the same as if(Math.random() <= 0.1), which means it will return early only 10% of the time, so the exception gets logged 90% of the time.

Wait, what?