When updating an Oracle Real Application Clusters (RAC) cluster with patches, it is generally recommended to follow a sequential patching approach rather than applying patches in parallel. Sequential patching involves patching one node at a time, while ensuring the cluster remains operational throughout the process.
Here are a few reasons why sequential patching is often preferred:
- Minimize downtime: Sequential patching allows you to minimize downtime by patching one node at a time while other nodes in the cluster continue to serve the workload. This approach ensures continuous availability of the cluster during the patching process.
- Mitigate risks: By patching nodes sequentially, you can identify any issues or errors that may arise during the patching process on a single node. If a problem occurs, it can be addressed before moving on to the next node, reducing the impact on the entire cluster.
- Easier rollback: If a patch causes unexpected issues or introduces instability in the cluster, it is easier to rollback or revert the changes on a single node rather than the entire cluster. This minimizes the potential impact on the overall system and simplifies troubleshooting.
- Testing and validation: Sequential patching allows you to test and validate the patched node before proceeding to the next one. This approach helps ensure that the patched node functions correctly and performs as expected, providing an opportunity to catch any issues early in the process.
Oracle RAC cluster patch update should be in parallel or sequential?
Right now we have parallel but Considering this is only meant for NON-Prod and we patch just after RAC creation, parallel is good for saving 40mins time . If its existing used cluster and the we should go for sequential patching allows us to minimize downtime by patching one node at a time while other nodes in the cluster continue to serve the workload. This approach ensures continuous availability of the cluster during the patching process.
But Best Approach in case new Cluster scenario:
This is an easy approach and most beneficial. Once we would update an patch manually and create an new AMI (Golden Image) and create a Oracle RAC cluster using new Golden Image only.
Benefit – X mins saving at during new cluster creation and exit status can be synced easily.
- DevOps Certified Professionals (DCP)
- Site Reliability Engineering Certified Professionals (SRECP)
- Master in DevOps Engineering (MDE)
- DevSecOps Certified Professionals (DSOCP)
URL - https://www.devopsschool.com/certification/
My Linkedin - https://www.linkedin.com/in/rajeshkumarin
- Best AI tools for Software Engineers - November 4, 2024
- Installing Jupyter: Get up and running on your computer - November 2, 2024
- An Introduction of SymOps by SymOps.com - October 30, 2024
