This document briefly describes the CloudMan/Galaxy flavors of CloudBioLinux available and how to configure them. For quick, small modifications to CloudMan your best bet is probably to modify an existing instance, documentation on how to do this can be found here. Documentation for building a CloudMan image and corresponding volumes and bucket from scratch using the CloudBioLinux deployer can be found here.
fab -f fabfile.py -i <key> -H ubuntu@<IP> install_custom:cloudman
A slightly more flushed out instance can be installed via the command:
fab -f fabfile.py -i <key> -H ubuntu@<IP> install_biolinux:flavor=cloudman/cloudman
Finally, CloudMan and Galaxy can be installed together via the command:
fab -f fabfile.py -i <key> -H ubuntu@<IP> install_biolinux:flavor=cloudman/cloudman_and_galaxy
You can additionally configure your CloudMan and Galaxy instance by specifying a configuration file: "-c <your_fabricrc.txt>" in the above command.
A subset of the parameters you may override via this mechanism includes (see
config/fabricrc.txt for a full list):
galaxy): User of Galaxy webapp
/mnt/galaxyTools/galaxy-central): Galaxy installation directory
/mnt/galaxyTools/tools): Galaxy tool's directory, formally called
/mnt/galaxyIndices): Directory for installation of Galaxy loc files.
https://bitbucket.org/galaxy/galaxy-central/): This is the mercurial repository of Galaxy to install into.
False): If this is
True, CloudBioLinux will not tweak the repository pulled in via mercurial for CloudMan. This is for applications which prepackage the needed changes to Galaxy directly into the Mercurial repository.
True): If this is
True, the Galaxy installation will be preconfigured for SGE and contain CloudMan branding.
There are two additional flavors assembled for the Galaxy-P project.
Below are some recommended fabricrc.txt overrides for Galaxy-P.
dist_name = precise
galaxy_repository = https://bitbucket.org/galaxyp/galaxyp-central/
galaxy_tool_conf = /path/to/cloudbiolinux/contrib/flavor/proteomics/galaxyp/galaxy_tools.yaml
Side Note: Galaxy-P is actively developed, tested, and deployed with Ubuntu 12.04 LTS, so this is what to target for best results. Feel free to let me (firstname.lastname@example.org) know if there are issues or if you would like help deploying in other environments and I will attempt to help in whatever way I can.
The Galaxy-P Server Flavor
This flavor is used to build the internal and public Galaxy-P servers hosted on the OpenStack cloud at the Minnesota Supercomputing Institute, though can of course be used to build your own Galaxy-P environment. This flavor is a sort of stripped down proteomics environment for Galaxy.
fab -f fabfile.py -i <key> -H ubuntu@<IP> install_biolinux:flavor=cloudman/cloudman_and_galaxyp
This flavor additionally requires a wine environment to be packaged using the
proteomics-wine-env project and made available to CloudBioLinux via the
setup_proteomics_wine_env_script fabricrc property.
The Galaxy-P Desktop Flavor
The Galaxy-P desktop flavor builds on the server flavor with additional desktop tools and programming libraries designed to make it a broadly useful environment for mass spec data analysis even outside the context of Galaxy while also stripping out the wine related tools to avoid potential legal concerns associated with redistributing such tools.
An Amazon AMI of this flavor will be available for launch on MSI's BioCloudCentral instance.
fab -f fabfile.py -i <key> -H ubuntu@<IP> install_biolinux:flavor=cloudman/cloudman_desktop_and_galaxyp