Permalink
a8b3837 Feb 17, 2017
@martenson @dannon @malloryfreeberg
251 lines (155 sloc) 13.4 KB
autotoc
true

Galaxy Download and Installation

In addition to using the main public Galaxy server (a.k.a. Main), you can install your own instance of Galaxy (what this page is about), create a cloud-based instance of Galaxy, or use one of the ever-increasing number of public Galaxy servers hosted by other organizations.

See Choices for help on deciding which of these options may be best for your situation.

Reasons to Install Your Own Galaxy

Installing your own Galaxy may be a good choice if you plan to:

Requirements

Get the Code

Galaxy's source code is hosted in a GitHub repository. Outlined below are basic options for obtaining the source code so you can use Galaxy. For more information see source code details.

For Production or Single User

If running a production Galaxy service or creating your own personal Galaxy server, use the latest release branch, which only receives stable code updates. The list of releases is available for viewing.

  • If you do not have a Galaxy repository yet or you do not want to update the existing instance, run:
$ git clone -b release_16.10 https://github.com/galaxyproject/galaxy.git
  • If you have an existing Galaxy repository and want to update it, run:
$ git checkout release_16.10 && git pull --ff-only origin release_16.10

For Development

To obtain Galaxy for development, use the default branch after cloning: dev. This is the branch that pull requests should be made against to contribute code (unless you are fixing a bug in a Galaxy release).

$ git clone https://github.com/galaxyproject/galaxy.git

Start It Up

Galaxy requires a few things to run: a virtualenv, configuration files, and dependent Python modules. However, starting the server for the first time will create/acquire these things as necessary. To start Galaxy, simply run the following command in a terminal window:

$ sh run.sh

This will start up the Galaxy server on localhost and port 8080. Galaxy can then be accessed from a web browser at http://localhost:8080. After starting, Galaxy's server will print output to the terminal window. To stop the Galaxy server, use Ctrl-c in the terminal window from which Galaxy is running.

To access Galaxy over the network, modify the config/galaxy.ini file by changing the host setting to

host = 0.0.0.0

Upon restarting, Galaxy will bind to any available network interfaces instead of the loopback.

You now have your very own Galaxy instance running - congratulations!


What to do next

Become an Admin

To control Galaxy through the UI (installing tools, managing users, creating groups, etc.), users must become an administrator. Only registered users can become admins. To give a user admin privileges, complete the following steps:

  • Add the user's Galaxy login email to the configuration file config/galaxy.ini. If the file does not exist, copy it from the provided sample: cp config/galaxy.ini.sample config/galaxy.ini.
  • Restart Galaxy after modifying the configuration file for changes to take effect.
  • Additional details can be found here.
# this should be a comma-separated list of valid Galaxy users
admin_users = user1@example.com,user2@example.com

Install More Tools

Galaxy comes with basic tools pre-installed. To install additinoal tools, follow the instructions on Installing tools into Galaxy from the ToolShed.

Join the Mailing List

To stay up-to-date on new Galaxy features and bug fixes, as well as to discuss future features, consider joining the Galaxy Developers mailing list. See Mailing Lists for other options.

Keep your instance backed up

Like any other application, Galaxy directories and Galaxy database tables should be backed up, and any disaster recovery plans should be regularly tested to make sure everything is working as expected.

Configure for production

The above instructions are intended for users wishing to develop Galaxy tools and Galaxy itself. To deploy a production-ready installation of Galaxy, some changes from the default configuration are highly recommended. If nothing else, switching to PostgreSQL or MySQL (from the default SQLite) is heavily endorsed to prevent database locking issues that can arise with multiple users.

Please see the Running Galaxy in a production environment page for more details.

Keep your code up to date

Galaxy development occurs in GitHub. Changes are stabilized in the release_YY.MM branches and then merged to master for each YY.MM.point release.

To be made aware of new Galaxy releases, please join the Galaxy Developers mailing list. Each release is accompanied by release notes.

At any time, you can check to see if a new stable release is available by using the git log command:

$ git log ..origin/master
commit 3f314974c9c3742b118518881a6d392123ccc05d
Merge: d8eeaae c78b7b6
Author: Nate Coraor <nate@bx.psu.edu>
Date:   Mon Mar 9 22:26:54 2015 -0400

    Merge branch 'release_15.03' to master for v15.03

If git log produces no output, Galaxy is up-to-date. If git log produces a list of commits, a new version is available. We suggest checking the accompanying release notes first (if the release is to a newer major version of Galaxy), but you can also immediately pull the commits to your local Galaxy clone with:

$ git pull

Note: After pulling changes, the Galaxy server needs to be stopped and restarted with the updated code. Restarting will interrupt any running jobs unless you are using a cluster configuration. For more information on how to make Galaxy restartable without interrupting users, see the production server documentation.

Note: Occasionally, updated Galaxy code includes structural changes to the database tables. The news brief will alert you if a release contains a database change. After updating and restarting Galaxy, Galaxy will refuse to load and will output an error message indicating that your database is the wrong version. The error message indicates that you should backup your database and run sh manage_db.sh upgrade. Follow those instructions carefully, especially the part about backing up your database safely! Database updates are carefully tested before release, but it is good practice to be able to back out if something goes wrong during an update.

In the unlikely event that something goes wrong with updated code, you can return to an older release by using the release tag name from the news brief page and the git checkout command. For example, to return to the latest version of the January 2015 release, use:

$ git checkout release_15.01

You can also use tags to check out specific releases:

$ git tag
v13.01
v13.01.1
v13.02
 ...
v14.10.1
v15.01
v15.01.1
v15.01.2
v15.03

Restore the fresh backup if a database update was required, and then restart Galaxy to get back to where you started.

Troubleshooting

  • Offline start: The initial Galaxy run requires Internet access to download the pre-built Python wheels of Galaxy's dependencies.

  • The basic Galaxy install is a single-user instance and is only accessible by the local user. As with many web-based applications, enable cookies in the web-browser for full functionality.

  • A common practice when using any web browser is to stay current with software updates to maximize performance and security. If moving forward to production server with login enabled, please make sure you and your end-users are current.

  • Some tools shipped with Galaxy have dependencies that need to be satisfied manually. Please see details here.

Help

Get Galaxy without Git

  • Galaxy from an archive
    • If you don't have Git (and thus cannot run the git command), you can download Galaxy in an archive instead: zipped or tar/gzipped.
    • Be aware that using archives makes it more difficult to stay up-to-date with Galaxy code because there is no simple way to update the copy.

Get Galaxy for Development

  • Fork Galaxy in GitHub
    • If you're doing development or making changes to Galaxy, it is best practice to fork Galaxy in GitHub and update to/from your fork.
    • See the GitHub fork documentation for details.

Become an Admin: Example

Follow these steps to become an admin for a brand new Galaxy.

Prerequisites

  • Galaxy was installed using Git.
  • Galaxy was started at least once using sh run.sh.
  • Galaxy is currently stopped. See our documentaion for help stopping Galaxy.

Check the default settings

The default setting for admin_users is: #admin_users = None. You can check this by running the following command:

$ grep "admin_users" config/galaxy.ini.sample

Create a config/galaxy.ini file and add yourself as an administrator

The following command will: 1) remove the leading hash (#) character from the admin_users line, 2) replace None with the email address of the user being added as an admin, and 3) create a config/galaxy.ini file with admin_users changed. Be careful to type the command exactly as written except for changing admin@email.edu to be the user's email address.

$ sed 's/#admin_users = None/admin_users = admin@email.edu/' config/galaxy.ini.sample > config/galaxy.ini

Confirm that the new config/galaxy.ini file is correctly formatted

Running the command below should result in the following output: admin_users = admin@email.edu

$ grep "admin_users" config/galaxy.ini

That's it for the galaxy.ini file changes!

Start up Galaxy again and create an account - using the same email address - through the Galaxy web interface (if this hasn't been done already). The Admin tab will now appear in the masthead for that account.

Shutting down Galaxy

Below are simplified instructions for shutting down local Galaxy server. If your configuration is more complicated, getting help from an administrator is recommended.

The Galaxy process is running in the background

  • If Galaxy was the last process running within a terminal window, bring it into the foreground with the command fg and shut down with Ctrl-c.
  • If Galaxy is one of many processes running in the background within a terminal window, find it with the command jobs. The list of jobs will be numbered. Bring the Galaxy job to the foreground with the command fg <number_of_the_job> and shut down with Ctrl-c.

I have lost the terminal window running Galaxy

  • From another terminal window, find all active processes with the command ps. The list of processes will each have a process ID (called PID). The target process will be named python ./scripts/paster.py serve config/galaxy.ini.sample. Stop it with the command kill PID, where "PID" is the actual process ID number.
  • If you kill only the process named sh run.sh, this will result in conflicts and Galaxy will not restart. If you did this or are simply getting errors when trying to restart Galaxy, the solution is to kill the process python ./scripts/paster.py serve config/galaxy.ini.sample before restarting Galaxy again.

Other Resources

  • Search all Galaxy administration resources.
  • Look into Galaxy KickStart Ansible playbook. To deploy Galaxy on your local machine or on the cloud using Ansible, you can use this playbook. Follow the steps in the docs for details.