End-to-End CI/CD Pipeline using Azure DevOps and Terraform for Oracle Cloud Infrastructure (OCI)

 Learn how to build a complete CI/CD pipeline using Azure DevOps and Terraform to automate Oracle Cloud Infrastructure (OCI). This step-by-step guide covers repository setup, service connections, Terraform configuration, pipeline creation, self-hosted agents, and deployment strategies. Ideal for DevOps engineers and cloud architects working with OCI, Azure DevOps, and Infrastructure as Code (IaC). Improve automation, scalability, and deployment efficiency with real-world implementation insights.

Introduction

In today’s cloud-native world, automation is the backbone of scalable infrastructure. In this blog, I’ll walk you through setting up a complete CI/CD pipeline using Azure DevOps and Terraform to deploy resources on Oracle Cloud Infrastructure (OCI).

This approach enables:

• Infrastructure as Code (IaC)
• Automated deployments
• Version-controlled infrastructure
• Scalable and repeatable environments

Create Azure DevOps Repository

Create a PAT token in Azure

Go to your profile and 

Clone the blank repository to your local machine  using an IDE. I had used VsCode

Download the terraform extension in Azure.

Create service connection to OCI

Choose OCI for terraform

Save the details

Prepare the terraform codes

Note: The terraform state file will be stored in the OCI object storage.

Create the variables in ADO

Click on pipelines. Inside we have the library

Also upload the private key to the secure files which will be referred during the execution

The local machine should have the OCI-CLI installed and configured so that terraform connects to the target OCI tenancy. 

Github Repo:-  https://github.com/samratsinha/ADO-OCI-Terraform

Push the code to ADO Repo

in ADO

Create the ADO Pipeline

Save it

Trigger the CI/CD Pipeline

Every commit in the code will trigger the ADO pipeline.

In Azure

in OCI

From the screenshot above, we can see that the pipeline completed successfully and created all the resources in OCI needed to run a VM. This concept can be scaled up as per the requirement.

In continuation to this blog, the part 2 of this blog will discuss about how we can add an another stage to the pipeline which is terraform destroy which will only run after the approval is provided in the staging pipeline.

Till then, continue learning cloud. Hope this will help someone.

How to Set Up Vulnerability Scanning Service in OCI (Step-by-Step Guide)

 

Introduction

Securing your cloud infrastructure is critical, and Oracle Cloud Infrastructure (OCI) provides a built-in Vulnerability Scanning Service (VSS) to detect security risks in your compute instances. In this guide, we’ll walk through the complete setup—from scan recipes to remediation.

---

Prerequisites

• A running OCI compute instance
• Proper access to OCI Console
• Required permissions to configure security services

Create scan recipes

Click on create scan recipe

Go to the compute instance and Enable the VSS agent

Create target. In this example, i am taking a single compute instance as an example.

Configure Cloud Guard and click on recipes

change it to root

click on Detector Rules tab and make sure the below ones are enabled

from the home page, click on configuration and then create new targets

Once the target is created, then move to scan reports

Click on vulnerability report

Go to cloud guard >Alerts>problems

The next step would be to remediate the problems based upon the individual scenarios one by one.

Conclusion

OCI’s Vulnerability Scanning Service combined with Cloud Guard provides a powerful security layer. Regular scans and proactive remediation ensure your infrastructure remains secure and compliant.

How to Add MCP to VS Code and Import Oracle Database Skills (Step‑by‑Step Guide)

 Modern AI development isn’t just about what your agent is trained on—it’s about what skills you give it. Skills dramatically expand your agent’s abilities without forcing it to waste tokens researching or relying solely on its training data. By adding skills, your agent can perform real actions, automate tasks, and interact with your Oracle Database environment instantly and efficiently. In my last blog-https://samappsdba.blogspot.com/2026/01/oracle-sqlcl-mcp-oracle-19c-nlp.html, i have described on how to leverage MCP protocol using sqlcli & Claude client and bring AI to the 19c Database.

This guide walks through how to set up Model Context Protocol (MCP) in Visual Studio Code, connect SQLcl, and import powerful Oracle Database skills from the GitHub repository at https://github.com/krisrice/oracle-db-skills. The result is a fully AI‑enhanced Oracle development environment inside VS Code.

 What You Need Before You Start

• Latest Visual Studio Code

• Latest SQL Developer Extension

• SQLcl installed

• Oracle Database 26AI or compatible version

• Git installed and added to PATH

• A workspace folder for skill imports

Install VS Code and SQL Developer Extension

Install VS Code, then add the SQL Developer extension and select Trust Workspace & Install

Once it is installed we will see the icon in the bottom as

Install Oracle Database 26AI

If you don’t have a database yet, install Oracle Database 26AI Free on a VM or local machine.

Connect to the Database Using SQLcl

C:\Users\SamratSinha>cd C:\AI\sqlcli\sqlcl\sqlcl\bin

 

C:\AI\sqlcli\sqlcl\sqlcl\bin>.\sql /nolog

SQLcl: Release 25.4 Production on Sun Mar 22 15:44:36 2026

 

Copyright (c) 1982, 2026, Oracle.  All rights reserved.

 

SQL> conn -save oracle19c_mcptest -savepwd system/******@IP of DB:1521/ FREEPDB1

Name: oracle19c_mcptest

Connect String: 132.145.169.175:1521/ FREEPDB1

User: system

Password: ******

Connected.

SQL> CONNMGR show oracle19c_mcptest

Name: oracle19c_mcptest

Connect String: 132.145.169.175:1521/ FREEPDB1

User: system

Password: ******

SQL> CONNMGR test oracle19c_mcptest

Oracle AI Database 26ai Free Release 23.26.0.0.0 - Develop, Learn, and Run for Free

Connection Test Successful

Restart VS Code so the SQL Developer extension can detect the MCP server

Enable Copilot MCP Integration in VS Code

In VS Code:

1. Open Command Palette

2. Search MCP: List Servers

3. Start the MCP server

4. Approve the session when prompted

Choose MCP:List Servers and then start MCP

Connect to the desired LLM

connect to the target DB

Import the skills from Kris Rice Github Repo

Once skills are imported, your agent can reuse them repeatedly without needing to re‑query the internet or regenerate complex logic. This means:

• Far fewer tokens consumed

• Faster responses

• More accurate actions

• A smarter, more capable agent over time

By giving your agent high‑quality skills—like the Oracle Database skills from Kris Rice’s GitHub, you’re effectively upgrading its intelligence and reducing operational cost at the same time.