Preparing a Power9 management server

Software package installation and the management server must be prepared before deployment or upgrade IBM Storage Scale System.

The following steps apply to any IBM Storage Scale System deployment or upgrade in a new cluster environment where no management server exists.

  1. Log in to the Power9 management server by using the management IP, which the SSR set up by using the provided worksheet. The default password is ibmesscluster.
  2. Optional: Set up a campus or a public connection (interface enP1p8s0f2). The connection might be named 'campus'. Connect an Ethernet cable to C11-T3 ports on the P9 management server and your lab network to access the GUI or call home to IBM. The container creates a bridge to the management network, thus a campus connection is highly advised.
    Note: If you do not set up a campus or a public connection, you temporarily lose your connection when the container bridge is created in a later step.

    This method is for configuring the campus network, not any other network in the management server node. Do not modify T1, T2, or T4 connections in the system after SSR set these connections, and use the SSR method only to configure T1 and T2 (if you must change after SSR is finished). That includes renaming the interface, setting IP, or any other interaction with those interfaces.

    You can use the nmtui command to set the IP address of the campus interface. For more information, see Configuring IP networking with nmtui tool.

  3. Import the /etc/hosts file on the management server node. This file must contain the low-speed IP addresses for the management server and high-speed IP addresses for the cluster, FQDNs, and short names. The high-speed names must contain a suffix to the low-speed names (For example, essio1-hs (high-speed name) to essio1 (low-speed name)). 
    127.0.0.1 localhost localhost.localdomain.local localhost4 localhost4.localdomain4

## Management IPs 192.168.45.0/24
192.168.45.20 <MS1>.localdomain.local <MS1>
192.168.45.21 <IOnode1>.localdomain.local <IOnode1>
192.168.45.22 <IOnode2>.localdomain.local <IOnode2>
192.168.45.23 <protnode1>.localdomain.local <protnode1>
192.168.45.24 <protnode2>.localdomain.local <protnode2>

## High-speed IPs 10.0.11.0/24
10.0.11.1 <MS1>-hs.localdomain.local <MS1>-hs
10.0.11.2 <IOnode1>-hs.localdomain.local <IOnode1>-hs
10.0.11.3 <IOnode2>-hs.localdomain.local <IOnode2>-hs
10.0.11.4 <protnode1>-hs.localdomain.local <protnode1>-hs
10.0.11.5 <protnode2>-hs.localdomain.local <protnode2>-hs


## Protocol CES IPs
10.0.11.100 <protnode>_ces1.localdomain.local <protnode>_ces1
10.0.11.101 <protnode>_ces1.localdomain.local <protnode>_ces1
10.0.11.102 <protnode>_ces2.localdomain.local <protnode>_ces2
10.0.11.103 <protnode>_ces2.localdomain.local <protnode>_ces2
    Note:
    • The localdomain.local domain name is an example that you cannot use for deployment. Replace the domain name with a valid fully qualified domain name (FQDN) during the /etc/hosts setup. The same domain name must be used across all defined network subnets.Set the domain on the management server.
      hostnamectl set-hostname NAME

      NAME must specify the fully qualified domain name (FQDN) of the management interface (T1) on the management server. If additional names are required for campus or other interfaces, configure names as aliases. Do not use these names as the primary hostname displayed by the hostnamectl command.

      You can set up the management server FQDN manually or wait until prompted when the IBM Storage Scale System deployment binary is started. At that time, the scripts confirm the FQDN and you can make changes.

    • If you want to set up a supported IBM Storage Scale System with the Power9 management server, add new IBM Storage Scale System hostnames to the /etc/hosts file by using the same structure. For example, low-speed (management) and high-speed (cluster) IP addresses, FQDNs, and short names.
    • Do not use any special characters, underscores, or dashes in the hostnames other than the high-speed suffix; example, -hs.
      Note: If you have protocol nodes, add the protocol nodes to the commands provided in these instructions. The default /etc/hosts file has hostnames prt1 and prt2 for protocol nodes. You might have more than two protocol nodes.
  4. Extract the installation package.
    Note: Check the installed version by manufacturing in SSR worksheet. If a new version is available in Fix Central, complete the following steps:
    1. Replace the existing image in the /home/deploy directory with a new image.
    2. Remove the old .tgz file before you extract the package.
      cd /home/deploy

xz --decompress ess_7.0.1.x_0611-23.tar.xz
  5. Complete the following steps, before you start the container:
    1. Create /home/log and /home/backup.
    2. Ensure that Chrony is running. For more information, see Setting a time zone.
  6. Accept the license and install the accepted image.
    Important: Before you create a container, delete any preexisting container.
    ./ess_7.0.1.x_0611-23.tar.xz

    Example of the contents of the extracted installation package:

    ├── 70-persistent-net-ems.rules

├── classes

│   ├── essmgr.py

│   ├── essmgr_yml.py

│   ├── _init_.py

│   └── _pycache_

│   ├── essmgr.cpython-36.pyc

│   ├── essmgr_yml.cpython-36.pyc

│   └── _init_.cpython-36.pyc

├── ess_7.0.1.x_0611-23.tar.xz
├── essmgr

├── essmgr.yml

├── essmkyml

├── logs

│   ├── essmgr.yml_2022-05-16_18-16-28

│   └── essmkyml_2022-05-16_18-16-28_log


├── podman_rh8_ppc64le.tgz


├── Release_note.ess_7.0.1.x_0611-23.tar.xz.txt


├── rhels-8.8-server-extra.iso

├── rhels-8.8-server-ppc64le.iso

└── rhels-8.8-server-x86_64.iso
    In this step, you are prompted to provide these inputs:
    • Container name, which must not be in the /etc/hosts file or resolvable by using DNS.
    • Confirmation of the management server FQDN, which must match what is set for the management IP in the /etc/hosts file. If this value needs to be changed or set, you can change by using the essmkyml script. The essmkyml script is located in the extracted directory. For example, 
      /home/deploy/ess_7.0.1.x_0611-23_0611-23.tar.xz
  7. Import the nodes configuration to the management sever. This command determines the node information based on virtual product data (VPD) and also exchanges the SSH keys.
    essrun -N essio1,essio2,ems1 config load -p
    Where:
    essio1,essio2
    Specifies all nodes of the same type in the building-blocks.
    ems1
    Specifies a management server.
    -p
    Specifies the root password. The default password is ibmesscluster. Change the root password after deployment is complete.
    -N
    Specifies all the nodes of the cluster to update the hosts.yml file for BMC. If this command is run on the selected nodes only, for example, only on I/O nodes, and management server node, the hosts.yml file is overwritten with the latest nodes.

    After this command is run, you can use -N NodeGroup for future essrun steps (For example, -N ess_ppc64le). There are different node group names forIBM Storage Scale System 3000.

  8. Check whether nodes have any issues before deployment or upgrade. Review the output, make needed changes.
    essrun -N essio1,essio2,ems1 update --precheck