Release Notes for Nvidia Bright Cluster Manager 8.2-29

Release notes for Bright 8.2-29

== General ==
= Improvements=

  • Add CUDA 11.8 packages
  • Add CUDA 11.7 packages
  • Add mlnx-ofed57 packages for the Mellanox 5.7 OFED stack
  • Update mlnx-ofed56 to version 5.6-2.0.9.0
  • Update mlnx-ofed54 to version 5.4-3.5.8.0
  • Update mlnx-ofed49 to version 4.9-5.1.0.0
  • Update cuda-dcgm to version 2.4.6.1
  • Update python2-novaclient to version 11.0.1 to resolve an issue with running novaclient on FIPS-enabled systems

= Fixed Issues=

  • mlnx-ofed: Incorrect values for the LD* environment variables in the Ubuntu openmpi module file

== CMDaemon ==
= New Features=

  • Introduce new CMDaemon advanced configuration flag that will allow the use of the head node hostname instead of the default master value for the AccountingStorageHost and ControlAddr parameters in the slurm.conf file

=Improvements=

  • CMDaemon certificates are now generated with a start date of 1 calendar day before the issue date, instead of the Unix epoch
  • Decrease the timeouts for the CMDaemon service so that CMDaemon is stopped faster

=Fixed Issues=

  • CMDaemon crash when the Grid Engine exec hosts list contains a node unknown to CMDaemon
  • Add full support for multi-value http request parameters, which resolves an issue where the “CMDaemon ready” service is not able to handle a list of services by name
  • An issue where the oomkiller health check may not detect the OOM killer has run on RHEL8 compute nodes

== Node Installer ==
=Improvements=

  • Allow the node-installer to continue configuring IPMI after a failure to set username and password if the user already exists

=Fixed Issues=

  • An issue with getclientid script when using bonded Mellanox ConnectX4-Lx ethernet interfaces
  • An issue with the configure_ipmi.pl script not working when the user id is set to 0

== Cluster Tools ==
=Fixed Issues=

  • An issue with cloning the mysql database when using cmha dbreclone when a configuration file /root/.my.cnf with other mysql credentials exists

== Machine Learning ==
=New Features=

  • Deprecated ML package cm-chainer-py39-cuda11.2-gcc9
  • Introduced packages cm-cudnn8.2-cuda11.4
  • Introduced packages cm-cudnn8.4-cuda11.4