kbase/narrprereq repository overview

!!!! This readme is no longer accurate and needs to be updated !!!!!!

This directory contains a dockerfile used to build a docker container (http://www.docker.io/⁠) that contains a self-contained, running narrative instance. The Dockerfile has some built in assumptions about how/where it is run, mostly due to restrictions by Docker

1. It expects to be run out of a directory that contains a narrative directory (containing the a copy of the narrative repo) and a bootstrap directory containing a bootstrap repo. This is an effort to centralize the configuration of the python environment in the bootstrap module
2. The source.list file in this directory also needs to be in the same directory as the Dockerfile, this is used to specify the remote package repos for the apt-get commands.

It also expects that there is a running docker service on the current host.

You can create the container with the following command (run from this directory):

docker build -t sychan/narrative .

Feel free to replace sychan/narrative with whatever name you prefer to call the docker container

You can then run the container using:

docker run sychan/narrative

The dockerfile is configured to spin up an ipython notebook listening on the default port (8888) across all interfaces. It will come up password protected by default with the kbasetest user password. You can set a new password by replacing the --NotebookApp.password= parameter in the CMD string with a different sha1 hash ( use the IPython.lib.passwd() method to prompt for a password - the return value is the sha1 hash).

Docker will map the internal port 8888 to a high port on public interface of the hosting machine. Use the "docker ps" command to find the port mapping. Example: root@docker1:~/dockerstuff/kb_narrative# docker ps ID IMAGE COMMAND CREATED STATUS PORTS cf6d42d5ee8b sychan/narrative2:latest /bin/sh -c /bin/bash 28 seconds ago Up 27 seconds 49517->8888

In this example, you can see that port 49517 is mapped to port 8888 within the docker container. This means that instead of connecting to http://localhost:8888⁠ as you would on your typical ipython notebook, you would point your browser at http://external_ip:49517/ to access the notebook.