Release 17.05

• On AWS, Galaxy 17.05 is available
• An updated tool set, for the current list see this file
• Slurm job manager configuration changes to improve job performance, including the ability to queue up to 200,000 jobs
• Fixed remote logging via Loggly

Release 17.01.01

• Fix share clones of older Galaxy CloudMan instances that used shed_tool_conf_cloud.xml job conf file

Release 17.01

• Two updates to the galaxyFS, including Galaxy 16.07 and 16.10, both available
  from https://launch.usegalaxy.org/. This includes a completely new install
  of all the available tools.

• Important: Update Galaxy job configuration so several tools will run
  using 4 threads. This has an important implication in that if there are no
  worker nodes available with the required number of nodes, jobs for those
  tools will never run. The list of tools that have been configured to run this
  way is available here.

• Replaced volume-backed GalaxyIndices file system with a CVMFS-backed file
  system. Clusters will no longer
  create a volume for the indices FS but instead fetch the necessary data from
  a replica service. Indices available via this service are the same as the
  ones available on Galaxy Main. If you need additional reference genomes,
  use the Data Managers and those will be installed on your galaxyFS.

• A new machine image on AWS (ami-3be8cd2c in region us-east-1 and
  ami-9ba490e8 in region us-west-1) with root file system size of 50GB to
  accommodate appropriately sized cache (32GB) for the CVMFS. The image now
  supports AWS Enhanced Networking.

• Don't automatically disable the master from running jobs when autoscaling is
  enabled.

• Set cluster name as the Name tag for volumes.

• Accommodate AWS long resource IDs.

• Add ability to turn on SSL from start via user data (use_ssl: True).

Release 16.05

• Update galaxyFS to include Galaxy 16.04 release.
• Add a placeholder node to Slurm (with 64 CPUs). This allows Slurm to accept
  jobs that request more resources than are currently available, hence allowing
  Galaxy to submit jobs that will eventually get executed vs. leading to an error.
• Fix displayed Galaxy file system size and usage for cloned clusters. Before,
  the main page was showing the size of the transient file system instead of
  the Galaxy file system.

Release 16.03

• Update galaxyFS to include Galaxy 16.01 release.
• Add advanced autoscaling controls. A user can now specify how long to allow
  jobs to wait in the queue for, how many jobs to maintain in the queue and
  by how many nodes to scale.
• For AWS, added ability to specify volume IOPS (either via iops user data or
  by entering your IOPS number on launch.usegalaxy.org). This allows to go
  from previous 40-200 IOPS to max 20,000 IOPS now.
• On AWS, built a new AMI (ami-b45e59de) with support for EBS-optimized
  instances.
• For AWS, added ability to launch clusters into non-default VPC subnets via
  launch.usegalaxy.org.
• When sharing a cluster with a custom galaxyIndices file system snapshot,
  make sure it gets shared as well (thanks to @hackdna for reporting the bug).
• Ensure the PostStartScript does not run before all the other services have
  been added (thanks to @hackdna for reporting the bug).
• Several updates for starting ClouderaManager (thanks to @ddavidovic).

Release 15.12

• This is a minor update release.
• Update Galaxy to the Galaxy 15.10 release.
• Preload the GIE IPyhton Docker container onto the image for faster startup.
• Quit storing tags in the cluster config to prevent persistent_data.yaml
  becoming too big when supplied as user data to worker instances.
• Updated library versions in requirements.txt.
• Add a more direct method for locating the Nginx configuration directory
  (thanks to @MatthewRalston)
• Made file system archive extraction more resilient and synchronous.