Overview
Diving into the intricacies of Oracle Clusterware management, this comprehensive guide elucidates the meticulous steps involved in the process of deleting and adding RAC nodes within the Oracle environment.
Good Node : SERVER_B
Problematic Node : SERVER_A
GRID_HOME=/usr/local/oragrid/19.3.0.0
VIP Name of Problematic Node= SERVER_A-vip
Purpose
The purpose of this comprehensive blog is to provide DBAs, system administrators, and Oracle enthusiasts with a detailed and systematic guide to effectively manage RAC nodes within Oracle Clusterware. By addressing both the process of removing a problematic node and adding a new cluster node
Resolution
Follow Below steps to Delete/Remove and Adding node from good node.
Followed below steps for reconfiguring CRS on node SERVER_A
==============================================
Perform below on Problematic Node SERVER_A:-
As a GRID Owner
===============
cd /usr/local/oragrid/19.3.0.0/oui/bin
./runInstaller -updateNodeList ORACLE_HOME=/usr/local/oragrid/19.3.0.0 “CLUSTER_NODES=SERVER_A” CRS=TRUE -silent -local
/usr/local/oragrid/19.3.0.0/deinstall/deinstall -local
Below Step is optional, as the deinstaller will provide the script to deconfigure the Node
As a root Node (SERVER_A)
===============================
/usr/local/oragrid/19.3.0.0/perl/bin/perl /usr/local/oragrid/19.3.0.0/crs/install/rootcrs.pl -deconfig -force
As a root on SERVER_B
============================
Per MOS document 1262925.1, perform the following steps to remove/delete node from GI Clusterware.
Remove the Node from the Grid Cluster
As the root user on node SERVER_B, perform the following from the GRid_Home/bin directory
- Stop the VIP resource for the node “XOMT-ORA0” ( This is not needed, as the GI stack on the problematic node has been deconfigure)
# srvctl stop vip -i SERVER_A
- Remove the VIP for the node “SERVER_A” ( This is not needed, as the GI stack on the problematic node has been deconfigure)
# srvctl remove vip -i SERVER_A -f
- 3. Check the state of the environment and ensure the VIP for node SERVER_A is removed.
# crsctl stat res -t
- Remove node SERVER_A from the Grid Infrastructure/ Clusterware
# crsctl delete node -n SERVER_A
- As the owner of the GI Installation (grid) , perform the following to clean up the GI inventory on the remaining node (SERVER_B).
$ cd $ORACLE_HOME/oui/bin
$ ./runInstaller -updateNodeList ORACLE_HOME=/usr/local/oragrid/19.3.0.0 “CLUSTER_NODES={SERVER_B}” CRS=TRUE -silent
- As root, list the nodes that are part of the cluster to confirm the node required (SERVER_A) has been successfully removed and
the only remaining node is node SERVER_B.
$ olsnodes -t -n
Per MOS document 1332451.1, perform the following steps to add node/instance in Oracle Clusterware and RAC.
Adding a Cluster Node on Linux and Unix Systems
1 Ensure that Clusterware is successfully installed on at least one node of cluster.
- Verify the integrity of the cluster and node SERVER_A.
$ cluvfy stage -pre nodeadd -n SERVER_A -verbose
- Navigate to the /usr/local/oragrid/19.3.0.0/addnode directory on node SERVER_B and run the addnode.sh script using the following syntax.
$ ./addnode.sh -silent “CLUSTER_NEW_NODES={SERVER_A}” “CLUSTER_NEW_VIRTUAL_HOSTNAMES={SERVER_A-vip}” “CLUSTER_NEW_NODE_ROLES={hub}” -ignorePrereq -ignoreSysPrereqs
- Check that cluster is integratedand that the cluster is not divided into separate parts by running the following CVU command.
$ cluvfy stage -post nodeadd -n SERVER_A –verbose
NOTE:- Once successfully completed the adding node from good node, now we need add database instance from good node to reconfigured node.
As an oracle user run below step in good node (SERVER_B)
============================
Run below command to add database instance for reconfigured Node – (SERVER_A)
$ srvctl add instance -db ORCLDB -instance ORCLDB1 -node SERVER_A
Conclusion:
This blog concludes with an emphasis on mastering the intricate process of node management within the Oracle Cluster ware ecosystem. By addressing both node removal and addition processes, readers gain a profound understanding of maintaining the stability and efficiency of Oracle environments. With a step-by-step breakdown of each process, this guide serves as a valuable resource for DBAs and system administrators seeking to navigate the complexities of RAC node management.
