I am so completely stumped on this one. A week ago I upgraded to
Passenger 2.2.4 and subsequently realized that my production rails app
was no longer working properly, it was delivering the 500 error page.
After digging around a bit, I couldn't find any errors being logged to
either the Apache error log or my app specific production/server logs.
I then posted on the passenger google group about this issue, and was
told to downgrade to passenger 2.1.3 because the issue was not
reproducible. Downgrading fixed the issue (or so it seemed) because
the app was working once again. I made a few modifications to the app,
committed to git, pushed to the server, and restarted. All of a
sudden, the app wasn't working again.
At this point I decided to try out Nginx/Thin as a web server
(assuming it was the apache/passenger bridge making the silent
errors). I've spent most of today getting the thin/nginx stack
working, and now that it is (I've tested a static test site), I'm
still receiving this 500 page. This makes me wonder if it's something
I'm doing wrong in Rails, or, heaven forbid, a Rails bug that I've
discovered (doubtful). Also, Running my app in development on my
Macbook Pro works just fine. I've even tried creating a fresh rails
app from scratch on production, and it is producing the same silent
I am completely and utterly stumped. Just to reiterate, I cannot find
a single issue being logged to any access or error logs that would
clue me in. Please assist. This has even made me *think* about going
back to PHP.
My setup is as follows: Fedora core 9 (Slicehost VPS 256), Rails
2.3.2, Nginx 0.7.61, Thin 1.2.2, Apache 2.2.9, Passenger 2.2.4.