The documentation included here describes the Cray System Management (CSM) software, how to install or upgrade CSM software, and related supporting operational procedures to manage an HPE Cray EX system. CSM software is the foundation upon which other software product streams for the HPE Cray EX system depend.
The CSM installation prepares and deploys a distributed system across a group of management nodes organized into a Kubernetes cluster which uses Ceph for utility storage. These nodes perform their function as Kubernetes master nodes, Kubernetes worker nodes, or utility storage nodes with the Ceph storage.
System services on these nodes are provided as containerized micro-services packaged for deployment via Helm charts. Kubernetes orchestrates these services and schedules them on Kubernetes worker nodes with horizontal scaling. Horizontal scales increases or decreases the number of service instances as demand for them varies, such as when booting many compute nodes or application nodes.
This information is intended for system installers, system administrators, and network administrators of the system. It assumes some familiarity with standard Linux and open source tools, such as shell scripts, revision control with git, configuration management with Ansible, YAML, JSON, and TOML file formats, etc.
Introduction to CSM Installation
This chapter provides an introduction to using the CSM software to manage the HPE Cray EX system which also describes the scenarios for installation and upgrade of CSM software, how product stream updates for CSM are delivered, the operational activities done after installation for on-going management of the HPE Cray EX system, differences between previous release and this release, and conventions used in this documentation.
This chapter outlines how to set up default credentials for River BMCs and ServerTech PDUs, which must be done before the initial installation of CSM, in order to enable HSM software to interact with River Redfish BMCs and PDUs.
This chapter explains how to get the CSM product release, get any patches, update to the latest documentation, and check for any Field Notices or Hotfixes.
This chapter provides an ordered list of procedures to follow when performing an initial install or a reinstall of CSM software. See the separate “Upgrade CSM” chapter for upgrade procedures.
This chapter provides an ordered list of procedures which can be used to update CSM software that indicate when to do operational tasks as part of the software upgrade workflow. See the separate “Install CSM” chapter for initial install and reinstall procedures.
This chapter provides an unordered set of administrative procedures required to operate an HPE Cray EX system with CSM software and grouped into several major areas:
CSM Troubleshooting Information
This chapter provides information about some known issues in the system and tips for troubleshooting Kubernetes.
This chapter provides background information about the NCNs (non-compute nodes) which function as management nodes for the HPE Cray EX system. This information is not normally needed to install or upgrade software, but provides background which might be helpful for troubleshooting an installation.
This chapter provides documentation on the REST APIs of the services in CSM.
This chapter provides explanations of terms and acronyms used throughout the rest of this documentation.
See LICENSE.