[zatkin]

If Heroku is running some zombie instance of your app, what if you were to just kill the instance you control and let Heroku run the zombie instance? When your service goes fully down, you'll know they've killed the zombie instance, so you can then run a clean copy.

[badestrand]

Not being able to update your code plus your app will completely crash randomly at some point between now and eternity - sounds great!

[oneepic]

It's salvageable. Maybe spin up some kind of Unix watch that pings the zombie and deploys the new version. (but then, what happens if you get another zombie... crap)