Rocks v4.2 Beta is released for i386 and x86_64 CPU architectures.




Quick Start

  1. Boot your frontend with the kernel roll.
  2. At the top-right of the first user screen, click 'CD-based roll'
  3. Then supply the base+bio+ganglia+hpc+java+sge+web-server meta roll
  4. Repeat steps 2 and 3 for the kernel roll, OS roll disk 1, and OS roll disk 2.
  5. Then click 'Done'

The remainder of the install has the same general flavor of a Rocks 4.1 install.

New Features

  • Beta Release of Bioinformatics Roll
  • The Bio roll contains tools and utilities to facilitate Bioinformatics computation. Included tools: Blast from NCBI, MPI-blast from LANL, EMBOSS, hmmer, etc, along with programming utilities such as Biopython and Bioperl.

  • Graphical Installer
  • The frontend and compute nodes now utilize Red Hat's graphical mode of the anaconda installer.


  • OS Roll based on CentOS release 4/update 3 and all updates as of June 28 2006.
  • Updated SGE roll to SGE 6 update 8.
  • Included tentakel as a cluster-fork alternative.
  • Auto partitioning now creates a /var file system to protect the frontend against services that log many messages and fill up the root file system.

Bug Fixes

  • ganglia no longer logs all errors /var/log/messages. This addresses a web-based usage mode of ganglia that sends frequent error messages to /var/log/messages.
  • Can once again access MySQL database via frontend's web site.
  • Can once again print labels for cluster nodes via frontend's web site.
  • In SGE Roll, included configuration code to cleanup MPI programs after user executes a 'qdel' command.
  • Now using native 'useradd' command. This address the issue useradd breaking when up2date refreshes the 'shadow-utils' RPM.
  • The '--ghost' flag was removed from autofs.
  • Permissions for the directories that hold the rolls are now always set to 755 every time rocks-dist is run.
  • During a compute node installation, if the compute node is connected to a public network (e.g., eth1) and if that public network has a DHPC server, the compute node will now ignore that DHCP server and continue to send DHCP requests until it hears from a rocks frontend.

Known Issues

  • To add a user to the cluster, you must run:
    1. useradd <username>
    2. rocks-user-sync
    3. tentakel '411get --all'

    We're investigating why the new user information is not sent to the compute nodes.

  • Using '' for frontend central installations does not work.