Supported operating system

Verify that the operating system is supported by DPOD as described in Hardware and Software Requirements. After the server OS is installed, this can be verified using the following command:

 cat /etc/redhat-release

Resources allocation

Allocate resources according to the chosen deployment profile as listed in Hardware and Software Requirements. After the server OS is installed, this can be verified using the following commands:

free -h
lscpu

Network requirements

Ensure you have at least one network interface installed and configured with full access to network services, such as DNS and NTP.
Some configurations, such as the Cell environment, require 2 network interfaces. See Network Preparation and Firewall Requirements for more details.

Root access

The installation must be performed by a root user. You cannot use sudo instead.
Do not override the PATH variable with a fixed value during login sequence, as this will override the value set by DPOD installation in .bash_profile and will cause various scripts to fail.
Do not use script command during the login sequence to make typescript of the terminal session for audit, as this will cause various scripts to hang.
Do not use trap command to clear the terminal on session close, as this will cause various scripts to get extra characters as their input and fail.
Do not print a disclaimer in .bashrc, as this will cause various scripts to get the disclaimer as their input and fail.

Disks, mount points, file systems and logical volumes

DPOD requires at least 3 disks (LUNs / physical / virtual) for the operating system, for the application/logs, and for the data. Some configurations, such as the Cell environment, require additional disks.
Please allocate the mount points / file systems on the different disks, as described in Table 1 below. It is strongly recommended to use logical volume manager (LVM) - particularly for the data disks.
Once configured, you may verify the configuration using the following command:

lsblk

• Tip: to create the mount points / file systems during RHEL installation:

  • Choose Installation Destination option.

  • Select all Local Standard drives and choose option "I will configure partitioning" under the "Other Storage Options" section.

  • Follow the table below and add all mount points with required definitions using the "+" button.

  • To create a volume group (sys, app, data), when applicable, open the "Volume Group" listbox and choose "create new volume group ...".

• Tip: To use LVM in AWS EC2 instances with RHEL 8.x and EBS disks, first execute dnf install lvm2 to install the LVM package, and use gdisk to create a partition. For more information, see https://aws.amazon.com/premiumsupport/knowledge-center/create-lv-on-ebs-partition/.

Store service dedicated OS user and group

The Store service requires a dedicated OS user and group to run. Consider executing the following command:

groupadd storeadms && useradd -g storeadms -md /home/storeadm -s /bin/bash storeadm

OS locale

The supported OS locale is en_US.UTF-8. Check the OS Locale Configuration and change it if necessary.

SELinux configuration

Changes in SELinux configuration might be needed. If SELinux is enforced on the DPOD server, please review possible required configuration changes.

Setup DNS

Setup DNS - your network admin may need to assist you with this action.

Using yum on RedHat

For RedHat only: Your system might need to be registered and subscribed to the Red Hat Customer Portal to be able to install all prerequisites using yum.
Registration and subscription may differ between organizations and RHEL version, so use the following commands just as a demonstration:

subscription-manager register
subscription-manager attach --auto

• For RHEL 7.x

  Code Block
  subscription-manager repos --enable=rhel-7-server-rh-common-rpms subscription-manager repos --enable=rhel-7-server-optional-rpms

• For RHEL 8.x

  Code Block
  subscription-manager repos --enable rhel-8-for-x86_64-baseos-rpms subscription-manager repos --enable rhel-8-for-x86_64-appstream-rpms

Setup NTP

Setup NTP - it has to be the same one configured in your IBM DataPower Gateways.

• Consult your Linux and network admin about the proper way to configure this service.

• For RHEL 7.x, ensure the NTP RPM is installed. Consider executing the following commands:

  Code Block
  yum install ntp ntpdate <ntp server hostname> systemctl enable ntpd.service systemctl start ntpd.service

• For RHEL 8.x, ensure the Chrony RPM is installed. Consider executing the following commands:

  Code Block
  yum install chrony chronyd -q 'server {ntp_server_name} iburst' systemctl enable chronyd.service systemctl start chronyd.service

Setup hosts file

Verify that the /etc/hosts file includes an entry with your server name mapped to your external server IP.
To find your server name, you may execute the command hostname.

Required RPMs

Verify the existence of the following RPMs from the official RedHat/CentOS yum repositories:

• httpd version 2.4.6-67 and above (together with the following dependencies: mailcap, apr, httpd_tools)

• mod_ssl

• mod_proxy_html

• curl

• wget

• unzip

• iptables

• iptables-services

• bc

• fontconfig

• squashfs-tools (make sure squashfs module is loaded - see more at https://access.redhat.com/solutions/5477831 - and that it is not disabled in /etc/modprobe.d)

• numactl

• pciutils

• nvme-cli

The installation is usually performed by executing yum. If the command fails to find the packages, you should manually download the RPM files and install them.

yum install -y httpd
yum install -y mod_ssl
yum install -y mod_proxy_html
yum install -y curl
yum install -y wget
yum install -y unzip
yum install -y iptables
yum install -y iptables-services
yum install -y bc
yum install -y fontconfig
yum install -y squashfs-tools
yum install -y numactl
yum install -y pciutils
yum install -y nvme-cli

The following RPMs are recommended for system maintenance and troubleshooting, but are optional: telnet client, net-tools, iftop, tcpdump

Ensure the httpd service is enabled and started by executing the command:

systemctl enable httpd.service && systemctl start httpd.service && systemctl status httpd.service

Cleanup

In case you are using yum, it is recommended to clean its cache to make sure there is enough space in /var (yum cache can take a lot of the space there). To clean yum cache, execute the command:

yum clean all

biosboot

sys (sda)

2

Standard Partition

BIOS BOOT

swap

sys (sda)

8192

LVM

swap

/boot

sys (sda)

2048

Standard Partition

XFS

/boot/efi

sys (sda)

200
(for UEFI installations for GPT partition)

Standard Partition

EFI System Partition

/

sys (sda)

8192

LVM

XFS

/var

sys (sda)

8192

LVM

XFS

/tmp

sys (sda)

4096
(recommended 16384)

LVM

XFS

/shared

app (sdb)

512

LVM

XFS

/app

app (sdb)

8192

LVM

XFS

/app/tmp

app (sdb)

4096

LVM

XFS

/installs

app (sdb)

11264

LVM

XFS

/logs

app (sdb)

12,288
(can be on other fast disk - preferred locally)

LVM

XFS

/data

data (sdc)

As described in Hardware and Software Requirements or according to the sizing spreadsheet in case one was provided by DPOD support team. Minimum of 100GB.

LVM

XFS

[Required only for cell members]
/data2, /data22, /data222, /data3, /data33, /data333, /data4, /data44, /data444

Dedicated 6 or 9 disks

Only for cell members, according to the sizing spreadsheet provided by DPOD support team. See Setup a Cell Environment for information about these disks/mount points.

LVM

XFS