Setting up Galaxy 15.10 or later

Note:

These instructions have been tested with Galaxy version 15.10 & 16.01

Preparations

Install and configure a Postgresql server.

Install and Configure Galaxy

Prepare Postgresql Database

Clone the galaxy repository that is currently in use by CloudMan. Make sure to choose the right release (See https://wiki.galaxyproject.org/Develop/SourceCode):
```
git clone https://github.com/galaxyproject/galaxy
git checkout release_16.01
```
###Note:

To update an existing galaxy instance simply git checkout release_XX.XX and then sh manage_db.sh upgrade

CloudMan currently (Mar 28, 2016) requires release_16.01.
Create a PostgreSQL database for use with Galaxy:
```
$ createdb galaxy
```

Create a user account for the galaxy database:

$ psql galaxy
CREATE USER galaxy WITH PASSWORD 'galaxy';
GRANT ALL PRIVILEGES ON DATABASE "galaxy" to galaxy;

Configure Refinery Settings

As of commit: https://github.com/parklab/refinery-platform/commit/d9d6d81b8d834daff5763d9392fe6f85d2be0f25 there is a setting: REFINERY_GALAXY_ANALYSIS_CLEANUP that allows you to specify how/when galaxy libraries, histories, and workflows should be deleted/purged.

The default option is: "ON_SUCCESS" which will perform deletion upon a successful Analysis.

Other available options are: "ALWAYS" and "NEVER".

Configure Galaxy Settings

The following steps assume that you are in your Galaxy's root directory

Create a directory named tool-dependencies.
```
$ mkdir tool-dependencies
```
Copy the sample configs:

cp config/galaxy.ini.sample config/galaxy.ini
cp config/tool_sheds_conf.xml.sample config/tool_sheds_conf.xml

Open galaxy.ini in a text editor.
Locate database_connection, uncomment the line and set database_connection = postgres://galaxy:galaxy@localhost:5432/galaxy.
Locate tool_dependency_dir, uncomment the line and set tool_dependency_dir = <BASEPATH>/galaxy/tool-dependencies, where <BASEPATH> is the absolute path to your galaxy directory.
Locate host, uncomment the line and set host = 0.0.0.0.
If the purging of histories is desired, locate allow_user_dataset_purge uncomment the line and set allow_user_dataset_purge = True
Optional, to help with debugging job failures: cleanup_job = onsuccess

Launch Galaxy and Create Account

Run sh run.sh in the root directory of the galaxy repository.
Go to http://127.0.0.1:8080 and create a new Galaxy user account using admin@example.com as email address and test123 as password.
Go back to galaxy.ini, locate admin_users, uncomment the line and set admin_users = admin@example.com.
Restart Galaxy.
Log into your Galaxy instance with your admin user account and create an API Key in the User menu.

Install Sample Tools and Workflows

Uncomment the test toolshed located in: config/tool_sheds_conf.xml.sample
Copy config/tool_sheds_conf.xml.sample into config/tool_sheds_conf.xml
Open http://127.0.0.1:8080 with your web browser.
Log in with your user and password.
In the top menu bar click Admin -> Search Tool Shed -> Galaxy test tool shed
see https://wiki.galaxyproject.org/Tool%20Shed)
Search for refinery and click refinery_test -> Preview and install -> Install to Galaxy
Optionally specify a new category - e.g. Refinery Platform and hit Install
Click Admin in the top menu bar and Manage Installed tools
Click on 'refinery_test' then proceed to click on the test workflows and install each one by clicking on Repository Actions -> import workflow to Galaxy
When you switch back to the home view of Galaxy, you should now see your installed tools in the left navigation bar category as well as the uploaded or imported workflows at Workflows.

Configure Host Environment for Your Galaxy Instance

Note

These instructions assume that the user has background knowledge in creating environment variables for their respective OS.

Optional: Create an environment variable called REFINERY_VM_TRANSFER_DIR and set it to a directory of your choice. This directory will be mapped into the VM and can be used to transfer data between the host and the VM and vice versa, e.g. for bulk import of datasets that are residing on the host into your Refinery instance.
Restart your VM with vagrant reload to make the above changes effective.

Pro-Tip: If you don't want to pollute your global environmental space, you can utilize virtualenv's postactivate hook ($VIRTUAL_ENV/bin/postactivate).

Configure Refinery and Import Workflows

Configure Refinery for Your Galaxy Instance

NOTE: You will not need the port 8080 as shown in these instructions if you are configuring this for a cloudman instance. Simply http://<master_node_private_ip> will suffice

Log into the Refinery Admin interface and create a new Instance in the galaxy_connector app:

Base URL: http://192.168.50.1:8080
Data url: datasets
Api url: api
Api key: <APIKEY>
Description: admin account

Troubleshooting: If you have problems importing workflows using ./manage.py import_workflows because the importer doesn't seem to be able to connect to Galaxy then your gateway's address might be wrong. To find out what the actual address of your gate way is run netstat -rn. This will print something like this:

Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
0.0.0.0         10.0.2.2        0.0.0.0         UG        0 0          0 eth0
10.0.2.0        0.0.0.0         255.255.255.0   U         0 0          0 eth0
192.168.50.0    0.0.0.0         255.255.255.0   U         0 0          0 eth1

Look for the entry with the destination 0.0.0.0 and copy the Gateway address. Next you have to edit your galaxy_connector instance's base URL. For example http://192.168.50.1:8080 --> http://10.0.2.2:8080.

Adding a Galaxy instance & Creating a WorkflowEngine

Before you can import Workflows from a Galaxy installation into Refinery, the following requirements have to be met:

You have to add a Galaxy Instance for the Galaxy installation in question to Refinery through the admin UI.
You have to create a Workflow Engine for this Galaxy Instance using the create_workflowengine command, which requires a Galaxy Instance id and the name of a group that should own the workflow engine, e.g. “Public”.
Galaxy Instance ids can be found here: http://192.168.50.50:8000/admin/galaxy_connector/instance/

$ manage.py create_workflowengine <instance_id> "<group_name>"

Alternatively, you can also create a workflow engine through the admin UI, in that case, however, you have to manually assign ownership to the managers of the group that should own the workflow engine.

Annotate and Import Workflows into Refinery

Now you are ready to import workflows into Refinery and execute them on data stored in the Refinery data repository.
You will need to Annotate or Import your own workflows before being able to run an Analysis within Refinery.