4 ms·
Not claiming "weirdest" by reading the other comments - but this is one off the top of my head. Application A is multi-threaded, huge, hardly maintained. It al
by wink 9y ago
Not claiming "weirdest" by reading the other comments - but this is one off the top of my head.
Application A is multi-threaded, huge, hardly maintained. It also constantly crashes. Would've probably taken a few ace C++ programmers quite a while to iron them all out.
Solution? Write a custom watch-dog application that restarts it once it crashed. (This was on Windows, and a while ago. Being a *nix guy myself, I've no clue if something -available- like daemontools could've been used back then).
Let's say I didn't know whether to facepalm or stand in awe.
- ashleyn 9y agoSounds ridiculous until you realise they probably weren't given the proper time to ever fix it. What's better at that point? Restart it until it doesn't crash at a significant performance penalty, or just leave it broken?