This article describes maintenance tasks for the Makahiki database, and describes the process for starting and stopping the Makahiki web server.
Note
In this article, a “%” prompt indicates that a command can be performed from any working directory in the virtual machine. Directories are specified otherwise.
The provisioning script that was run when the virtual machine was started for the first time initialized the database. This section describes how to initialize the database again.
Warning
Running the initialize_instance.py script will:
- Install and/or update all pip-installed packages required by Makahiki.
- Reinitialize the database contents and perform any needed database migrations.
- Initialize the system with data.
- Set up static files.
This script should be run only a single time in production scenarios, because any configuration changes that an administrator makes will be lost if the initialize_instance.py script is used again.
The script initializes the Makahiki database and populates it with default information and users.
If the server is running in your terminal, close it:
(type control-c in the shell running the makahiki server process)
Change to the /vagrant/makahiki directory if you are not in it. (The % sign represents a command prompt and indicates that this can be done from any directory on the virtual machine):
% cd /vagrant/makahiki
Next, run the initialize_instance.py script:
vagrant@precise32:/vagrant/makahiki$ ./scripts/initialize_instance.py --type default
You will need to answer Y to the question Do you wish to continue (Y/n)?.
Example output of the initialize_instance.py script:
vagrant@precise32:/vagrant/makahiki$ ./scripts/initialize_instance.py --type default
installing requirements...
WARNING: This command will reset the database. All existing data will be deleted. This process is irreversible.
Do you wish to continue (Y/n)? Y
resetting the db...
DROP DATABASE
DROP ROLE
CREATE ROLE
CREATE DATABASE
syncing and migrating db...
collecting static and media files...
loading base data...
loading fixture base_badges.json...
loading fixture base_help.json...
loading fixture base_notifications.json...
loading fixture base_pages.json...
loading fixture base_quests.json...
loading fixture base_schedule.json...
loading fixture base_settings.json...
loading fixture smartgrid_library.json...
setting up default data...
set up 1 one-week rounds, starting from today.
loading fixture default_challenge.json...
loading fixture default_designer.json...
loading fixture default_prizes.json...
loading fixture default_smartgrid.json...
loading fixture default_teams.json...
0 test users deleted.
4 test users created.
event dates adjusted to round date.
created initial resource usages for all teams.
created test baselines for all teams.
created goal settings for all teams.
makahiki cache cleared.
vagrant@precise32:/vagrant/makahiki$
If the script experiences errors while connecting to the database, see Troubleshooting Makahiki on Vagrant.
Makahiki is designed to support post-installation updating of your configured system when bug fixes or system enhancements become available.
If the server is running in your terminal, close it:
(type control-c in the shell running the makahiki server process)
Change to the vagrant directory if you are not in it. (The % sign represents a command prompt and indicates that this can be done from any directory on the virtual machine):
% cd /vagrant
Download the updated source code into the Makahiki installation:
vagrant@precise32:/vagrant$ git pull origin master
Change your directory to vagrant/makahiki:
vagrant@precise32:/vagrant$ cd makahiki
Next, run the update_instance.py script:
vagrant@precise32:/vagrant/makahiki$ ./scripts/update_instance.py
This updates the Makahiki instance based on any new files that have been added to the Git repository.
Example output of the update_instance.py script:
vagrant@precise32:/vagrant/makahiki$ ./scripts/update_instance.py
installing requirements...
syncing and migrating db...
collecting static and media files...
vagrant@precise32:/vagrant/makahiki$
This guide assumes you are in the directory /vagrant/makahiki on the virtual machine.
Makahiki’s manage.py script provides two web servers: runserver and gunicorn. runserver is better for development, and gunicorn is better for production use.
It is important to bind the server to IP 0.0.0.0 (accepts incoming connections on any IP address) and port 8000 in order to work with the port forwarding settings in the Vagrantfile.
To start the server with manage.py:
vagrant@precise32:/vagrant/makahiki$ ./manage.py runserver 0.0.0.0:8000
Example output of starting runserver:
vagrant@precise32:/vagrant/makahiki$ ./manage.py runserver 0.0.0.0:8000
Validating models...
0 errors found
Django version 1.4, using settings 'settings'
Development server is running at http://0.0.0.0:8000/
Quit the server with CONTROL-C.
To start the server with gunicorn:
vagrant@precise32:/vagrant/makahiki$ ./manage.py run_gunicorn -b 0.0.0.0:8000
Example output of starting gunicorn:
vagrant@precise32:/vagrant/makahiki$ ./manage.py run_gunicorn -b 0.0.0.0:8000
Validating models...
0 errors found
Django version 1.4, using settings 'settings'
Server is running
Quit the server with CONTROL-C.
2013-10-11 01:59:41 [1399] [INFO] Starting gunicorn 0.13.4
2013-10-11 01:59:41 [1399] [INFO] Listening at: http://0.0.0.0:8000 (1399)
2013-10-11 01:59:41 [1399] [INFO] Using worker: sync
2013-10-11 01:59:41 [1408] [INFO] Booting worker with pid: 1408
View the site in your host machine’s web browser at http://192.168.56.4:8000.
Log in with the username and password specified in makahiki_env.sh. The username is “admin” and the password is “admin” unless these settings were changed after installation.
To stop either of the servers, type control-c in the virtual machine terminal.
If you cannot reach the web server from the host machine, you will need to use wget to test the server on the virtual machine:
vagrant@precise32:/vagrant/makahiki$ ./manage.py runserver 0.0.0.0:8000 &
Validating models...
Development server is running at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
vagrant@precise32:/vagrant/makahiki$ (press "Enter" here)
vagrant@precise32:/vagrant/makahiki$ cd ~/
vagrant@precise32:~$ mkdir test
vagrant@precise32:~/test$ cd test
vagrant@precise32:~/test$ wget http://127.0.0.1:8000
--2013-08-09 11:19:25-- http://127.0.0.1:8000/
Connecting to 127.0.0.1:8000... connected.
HTTP request sent, awaiting response... 302 FOUND
Location: http://127.0.0.1:8000/landing/ [following]
[09/Aug/2013 11:19:26] "GET / HTTP/1.0" 302 0
--2013-08-09 11:19:26-- http://127.0.0.1:8000/landing/
Connecting to 127.0.0.1:8000... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [text/html]
[09/Aug/2013 11:19:26] "GET /landing/ HTTP/1.0" 200 6181
Saving to: “index.html"
[ <=> ] 6,181 --.-K/s in 0s
2013-08-09 11:19:26 (192 MB/s) - “index.html" saved [6181]
If your HTTP response is “200 OK,” the server is running correctly. You can delete the “test” directory when you are done.
Because the server was started in the background with &, you cannot stop it with control-c. You will need to find the PIDs of its processes first:
% ps ax | grep manage.py
21791 tty1 S 0:00 python ./manage.py runserver
21798 tty1 Sl 0:52 /root/.virtualenvs/makahiki/bin/python ./manage.py ru
nserver
21893 tty1 S+ 0:00 grep manage.py
% kill -9 21791 21798
%
[1]+ Killed ./manage.py runserver (wd: ~/makahiki/makahiki)
(wd now: <your-working-directory>)
The PID of a given process will be different each time it runs. kill -9 <PID> forces the OS to stop the process with the specified PID. Kill the python ./manage.py runserver and /root/.virtualenvs/makahiki/bin/python ./manage.py runserver processes to stop the server.