Base Command Manager / Bright Cluster Manager Release Notes
Release notes for Bright 6.1-4 (released 2013-06-14)
== cmdaemon ==
* Added: config noreplace on /cm/local/apps/cmd/scripts/nodereconnectcheck.sh
* Added: extra validation for mysql tables
* Improved: MIC network configuration
* Improved: make cmsh vpc aware when adding new cloud nodes
* Improved: reorder the network properties in CMSH -- swapped Dynamic Range Start/End fields, so that 'start' is before 'end'.
* Improved: server side validation of disksetup
* Fixed: shorewall stop not clearing the iptables
* Fixed: clear restart required message when doing 'open -e -n ...'
* Fixed: mysql update for clusters already upgraded from 6.0 to 6.1
* Fixed: headnode cmd restart not triggering config writer update of managed services
* Fixed: cloud director reported UP->DOWN->UP the first time after startup
* Fixed: PBS Pro and TORQUE node drain/undrain when no queue is assigned
== node-installer ==
* A recent update added functionality to allow swap priority to be specified in the disk layout. This update adds functionality to also include swap priority when generating XML from the current disk layout. (This is used during failover setup.) In addition this solves a bug where generating XML would miss any swaps on devices that did not also contain other mounted file-systems.
== manuals ==
* added: OpenACC introduction
* added: examples mpi non-blocking messaging
* added: login nodes with Torque 4
== buildmaster ==
* update Dell hardware vendor package selections (DTK 4.3, OM 7.3)
* Added gcc-recent to default packages list for SLES11.
== cluster-tools ==
* Improved cm-mic-setup
* cm-register-node: fix attribute name 'myhostname'
* Changed the default domain name of the cloud network to cloud.cluster
* Private Cloud creation wizard now asks for a base ip range for the VPC.
== cm-config-ssh ==
* Fixed: Users' ~/.ssh directory not being created with the proper permissions
== cm-config-ssh-slave ==
* Fixed: Users' ~/.ssh directory not being created with the proper permissions
== cmgui ==
* MIC: Add bridge kernel module to the images for MIC host, don't setup alias bridge interfaces
* Fixed: event acknowledgment
* Fixed: selecting all queues in workload managers
== cuda42 ==
* Changed /cm/local module paths, to be able to also use more up to date nvidia driver package.
* Fixed MANPATH for the toolkit
* Changed unloading gcc to conflict with gcc in module files. Unloading another module does not work with environment-modules version 3.2.10
* Updated verify scripts: unload gcc module before loading cuda/toolkit module, use more CPU cores for compilation, changed check for installed cuda packages.
== cuda50 ==
* Changed unloading gcc to conflicts with gcc in module files. Unloading another module does not work with env-modules version 3.2.10
* Updated verify scripts: unload gcc module before loading cuda/toolkit module, use more CPU cores for compilation, changed check for installed cuda packages.
== intel-wired-ethernet-drivers ==
* Update of ixgbevf driver from 3.14.5 to 3.15.1
== munge ==
* Add --force option to allow munge to start after a crash
== sge ==
* sge test script: add gcc to the module load list, required for gfortran compatibility on rhel5 based systems
== slurm ==
* Fixed: prolog script not being properly set up
== torque-slave ==
* Fixed: torque_mom not able to start on SLES11