hpi-swt2/event-und-raumplanung

[SHOWSTOPPER] Heroku Dev "Application Error"

Closed this issue · 10 comments

Some commit of this day seems to have created a major application error that occurs on any route at http://event-und-raumplanung-dev.herokuapp.com/

This error is not logged on errbit
We traced it down to be an issue after the commit 0943c86 and before the commit 3e466c9
To anyone with a commit in this range: @ArneMayer @hsgfan @hagerapante
Please revisit your created revisiions on your local machine and revert or fix as soon as possible
The sooner you do that the earlier you are off the hook... ;)

Nochmal die Scrum Master, der Form halber: @Atyanar @lulufaber @rosabraatz @DasDingGehtNicht @SvenLehmann

I don't see any errors. Is it solved or is there another problem? Could you elaborate the problem?

I don't get any errors either.

What the Heck?
So we all got that error, even after re-running Travis
no commit since then...

Has this ever happened before?

Sorry, my fault. I forgot resetting the dev to the faulty state. There it is again.

Okay, so the problem persists
Please review your changes

hubx commented

passiert so früh, das errbit nicht getriggert wird. hier die heroku logs. Weil ich immer noch sehr wenige contributors im Heroku sehe, hier die logs

**2015-01-24T12:36:24.723633+00:00 app[web.1]: /app/app/controllers/ical_controller.rb:1:in <top (required)>': uninitialized constant AppilcationController (NameError)** 2015-01-24T12:36:24.723657+00:00 app[web.1]: from /app/vendor/bundle/ruby/2.1.0/gems/railties-4.1.6/lib/rails/initializable.rb:30:in run'

AppilcationController

hubx commented

Wie kann denn sowas durch die Tests durchgehen... es gibt nur Tests für 6/22 Controllern :(?

Ihr könnt die Production environment auch lokal mit SECRET_KEY_BASE=xxx RAILS_ENV=production rails s testen. Da wäre euch der Fehler auch aufgefallen.

I pushed a fix for this directly to the dev. Please have a look at it @DasDingGehtNicht and also be aware in the other teams. @Atyanar @lulufaber @rosabraatz @SvenLehmann

Seems to be fixed, but we have a nice follow-up issue #312