Skip to main content
FlashGrid.io

Out of Place ( OOP ) Patching using opatchauto

  • Updated
  • Products: FlashGrid Cluster on Azure
  • Products: FlashGrid Cluster on GCP
  • Products: FlashGrid Server on AWS
  • Products: FlashGrid Cluster on AWS
  • Products: FlashGrid Server on Azure
  • Products: FlashGrid Server on GCP

This document explains additional steps that need to be performed in FlashGrid cluster/server environments following out of place (OOP) Patching.

Note: The instructions in this article assume FlashGrid Diags version 20.2.187 or newer. To check your version, run rpm -qa|grep flashgrid-diags.

Out Of Place Patch Prerequisite

FlashGrid deployments have swap disabled by design. This prevents brown-out scenarios in case of low memory. However, some Oracle utilities still check for swap enabled for historical reasons.

It is possible to create a swap file once and reuse it for future patching. If a swap file was already created during a previous patching process, you can skip the first step and start from the second step.

1. Create a temporary 1GB swapfile:

# dd if=/dev/zero of=/swapfile bs=1024k count=1024 oflag=direct
# chown root:root /swapfile
# chmod 600 /swapfile
# mkswap  /swapfile

2. Activate the swap file before patching

# swapon /swapfile

3. Follow the patching procedure and return back to this page

4. Deactivate the swap file

# swapoff /swapfile

It is important to note that you should not leave the swap file enabled after patching. The swap file should be activated prior to patching only and then deactivated afterward. Failing to deactivate the swap file can result in performance issues and instability in the system.

Grid Infrastructure Post Patch

Restart FlashGrid ASM Service

# systemctl restart flashgrid_asm.service

Modify User Profiles

Ensure that the Oracle Grid Infrastructure Home Path is updated in the following files:

  • /home/fg/.bashrc
  • /home/grid/.bashrc
  • /root/.bashrc

For example, if there is the original home path in one of these files:

export ORACLE_HOME=/u01/app/19.3.0/grid
export PATH=/u01/app/19.3.0/grid/bin:$PATH

Change it to the clone home path:

export ORACLE_HOME=/u01/app/19.3.0/grid_oop
export PATH=/u01/app/19.3.0/grid_oop/bin:$PATH

Modify /etc/oratab (only for Single-Instance Databases)

Note: Follow the steps in this section only if you use Single-Instance Databases.

There are two types of entries in /etc/oratab:

  • managed by Oracle Restart - they have the # line added by Agent comment: 
    +ASM:/u01/app/19.3.0/grid:N             # line added by Agent
orcl:/u01/app/oracle/product/12.2.0/dbhome_1:N          # line added by Agent
  • non-managed by Oracle Restart - those are without the comment 
    asm:/u01/app/19.3.0/grid:N
exampledb:/u01/app/oracle/product/12.2.0/dbhome_1:N

    When you have any non-managed by Oracle Restart entries, update them accordingly to point them to the clone home path:

    asm:/u01/app/19.3.0/grid_oop:N
exampledb:/u01/app/oracle/product/12.2.0/dbhome_oop:N

Database Post Patch

Note This section should be skipped when out of place Patching is applied only to Grid Infrastructure. Otherwise, follow the steps below.

Modify User Profiles

Ensure that the Database Home Path is updated in: /home/oracle/.bashrc

For example, if there is the original home path:

export ORACLE_HOME=/u01/app/oracle/product/19.3.0/dbhome_1
export PATH=/u01/app/oracle/product/19.3.0/dbhome_1/bin:$PATH

Change it to the clone home path:

export ORACLE_HOME=/u01/app/oracle/product/19.3.0/dbhome_oop
export PATH=/u01/app/oracle/product/19.3.0/dbhome_oop/bin:$PATH

Modify /etc/oratab (only for Single-Instance Databases)

The steps are the same as in Modify /etc/oratab (only for Single-Instance Databases)