Question

4

Horizon: Apache exceptions at import_module

rated 0 times [ 4] [ 0] / answers: 1 / hits: 972 / 3 Years ago, fri, october 29, 2021, 1:06:28

I've deployed Openstack using juju, and everything looks normal except that I cannot browse to openstack-dashboard. I get error 500 Internal Server Error.

I looked through apache2 logs at that node and found these errors repeated every couple of minutes.

I tried to destroy the service and deploy it on other nodes; co-located with keystone node, then glance node, then by it self. Yet I get the same errors.

It looks like it needs deep python debugging which is weird since this is the typical charm deployment on a normal server. And someone must had tried to deploy this and succeeded!

I'm deploying openstack Essex version 2012.1.4-dev ,
Ubuntu server version 13.10 ,
django version is: (1, 5, 4, 'final', 0)
And this is my juju status output

Please help me :)

Answers

Only authorized users can answer the question. Please sign in first, or register a free account.

iriousutur

Add To Favorites

Follow

Total Points: 321

Total Questions: 112

Total Answers: 106

Location: Sweden

Member since Mon, Dec 7, 2020

3 Years ago

answered 3 Years ago overine · Accepted Answer

I am not sure about Essex, but OpenStack Grizzly is not compatible with django 1.5(the version you are using); you need to have django 1.4 installed.

To uninstall django 1.5, open a terminal(Ctrl+Alt+t) and type the following:

sudo pip uninstall django

and then to install django 1.4:

sudo pip install django==1.4.10

confirm version 1.4 is installed:

python

you will get the python interpreter

>>> import django;django.VERSION

the output should be:

(1, 4, 10, 'final', 0)

then restart apache service(which uses django):

sudo service apache2 restart

Source: Similar question on OpenStack's official question-answer forum