I just created a new django app in the webfaction control panel, using the latest, python3.3 and django 1.6 pair.
I've tested via runserver locally (created initially via django 1.6's startproject) and runs fine. I've posted my app to server, and testing with python3 and the shell, and runserver the app appears to load and models import without issue. It's only when I try to actually access the page (no change after restarting apache) Checking my ~/logs/user/myapp_error.log I see the following:
The strange thing is that running a test wsgi hello world program gives me no issues:
I haven't added or changed anything in myapp/apache2/conf/httpd.conf other than to adjust the paths to myapp name. Note I have installed python3 libraries into ~/webapps/myapp/lib/python3.3, but like I said it's importing fine outside of apache. If I had some kind of traceback to work with I could probably figure this out, but I'm stuck. It seems that many mod_wsgi Segfault issues point to compilation linking errors with apache/mod_wsgi, but that's something I don't have control over... Any ideas on how to proceed with this would be helpful. asked 10 Dec '13, 08:36 monkut |
It's probably because you're on an older machine, running CentOS 5. You'll either need to stick with older applications or else migrate to a newer server. There's some work involved with migration, even for the automated migration option (specifically, updating apps), but there are also some significant benefits to the newer servers as well. answered 12 Dec '13, 02:58 ryans ♦♦ thanks! Yeah, sounds like the migration is a good idea. Better than trying to fight with the past.
(12 Dec '13, 08:05)
monkut
|
This sounds like the type of error you might see if running very new software against very old libraries. Are you on a CentOS 6 server, or one of the older servers?
just checked, looks like I'm on a centOS5 machine.