Base Command Manager / Bright Cluster Manager Release Notes

Release notes for Bright 8.1-18

== General ==

- New Features

* Update to latest patch release for Kubernetes and apply patch for CVE-2019-11247

- Improvements

* Support for a generic BMC indentify script

== cmdaemon ==

- New Features

* Added an option (set revision global-domain:no) to disable generic device on global network

- Improvements

* Data overload in monitoring can cause crash in cache
* Increase maximal JobInformationKeepCount from 1M to 1G

- Fixed Issues

* In some cases, an issue with DB update on the passive head node
* Timed and prejob checks are not run as pejob
* Rare memory leak when socket connections time out
* In some cases, restart required flag is not cleared after the head node has been restarted
* GetNodeStatus calls returning 0

== node-installer ==

- Fixed Issues

* In some cases with HA clusters with Direct Connect or Express Route setups, an issue with updating /etc/hosts for cloud nodes, which can lead to missing "master" entry

== cmsh ==

- Fixed Issues

* cmsh can hang when the table column size gets rescaled to 1 character
* cmsh job status field is not set

== slurm ==

- Improvements

* Upgrade slurm19 to 19.05.2