OIC Agent High Availability Guide

Introduction

Oracle Integration Cloud Agent High Availability is a critical topic for any enterprise implementing hybrid integrations using Oracle Integration Cloud. In real-world projects, especially in industries like banking, manufacturing, and healthcare, downtime in integrations can directly impact payroll processing, order fulfillment, or employee onboarding.

When we deploy the connectivity agent to connect on-premise systems with Oracle Cloud Infrastructure (OCI), ensuring high availability (HA) becomes non-negotiable. A single agent instance failure can halt integrations unless proper HA architecture is designed.

This blog explains how to design, configure, and test high availability for Oracle Integration Cloud Agent in a practical, consultant-driven approach, aligned with the latest OIC Gen 3 (26A) practices.

What is Oracle Integration Cloud Agent High Availability?

Oracle Integration Cloud Agent is a lightweight Java-based component installed on-premise to securely connect OIC with internal systems such as databases, ERP, or legacy applications.

High Availability (HA) in this context means:

Running multiple agent instances under the same agent group so that if one instance fails, others continue processing integrations seamlessly.

Key Concept

• Agent Group = Logical grouping
• Multiple Agents = Physical instances
• Load Distribution = Automatic across active agents
• Failover = Automatic when one agent is down

Key Features of OIC Agent High Availability

1. Active-Active Architecture

All agents in the group actively process requests.

2. Automatic Load Balancing

Requests are distributed across available agents.

3. Fault Tolerance

If one agent fails, others continue processing without manual intervention.

4. Scalability

You can add more agents as integration load increases.

5. Zero Downtime Maintenance

You can stop one agent for maintenance while others continue running.

Real-World Integration Use Cases

Use Case 1: Payroll Integration (HCM → On-Prem ADP System)

In an HCM implementation, payroll data from Fusion needs to be pushed to an on-prem payroll system.

• If the agent fails during payroll processing:
  • Salary processing gets delayed
  • Compliance issues arise

Solution: Deploy 2–3 agents in HA mode to ensure uninterrupted data flow.

Use Case 2: Manufacturing Order Processing (SCM → Legacy ERP)

• Orders from Fusion SCM are sent to legacy systems
• Continuous processing is required for production planning

HA Benefit: Even if one agent crashes, order processing continues without delay.

Use Case 3: Real-Time Employee Sync (HCM → Active Directory)

• Employee creation triggers account creation in AD
• Requires near real-time processing

HA Impact: Ensures onboarding workflows are not disrupted due to agent downtime.

Architecture / Technical Flow

High-Level Flow

1. OIC sends request to Agent Group
2. Agent Group distributes request to available agents
3. Agent processes request and returns response
4. If one agent is unavailable → request routed to another agent

Components Involved

• Oracle Integration Cloud (OIC Gen 3)
• Connectivity Agent
• Agent Group
• On-Premise Systems
• Secure HTTPS Communication

Architecture Design (Consultant View)

In a production environment:

• Minimum 2 agents (recommended 3)
• Installed on different servers or VMs
• Same agent group configuration
• Shared credentials and certificates

Prerequisites

Before setting up HA, ensure:

1. OIC Instance Ready

Provisioned in Oracle Cloud Infrastructure

2. Agent Group Created

Already configured in OIC

3. Network Configuration

• Outbound HTTPS (port 443) enabled
• No inbound firewall rules required

4. Java Installed

• JDK 11 or higher

5. Same Agent Configuration File

All agents must use the same group identifier

Step-by-Step Build Process

Step 1 – Create Agent Group in OIC

Navigation:

Integrator → Settings → Connectivity Agents → Create Agent Group

Example Values:

Click Save

Step 2 – Download Agent Installer

• Download connectivity agent from OIC console
• Extract ZIP file on each server

Step 3 – Configure First Agent Instance

Edit agent.properties file:

Run agent:

Step 4 – Configure Second Agent Instance

On another server:

Start the agent.

Step 5 – Verify Agent Registration

Navigation:

Integrator → Settings → Connectivity Agents

You should see:

Step 6 – Configure Third Agent (Recommended for Production)

Repeat the same process for:

Testing the High Availability Setup

Test Scenario

Create an integration:

• Trigger: REST API
• Invoke: On-prem database via agent

Test Steps

1. Run integration with all agents active
2. Stop one agent manually
3. Trigger integration again

Expected Results

Validation Checks

• No integration failures
• No connectivity errors
• Logs show traffic handled by active agents

Common Errors and Troubleshooting

1. Agent Not Appearing in OIC

Cause: Incorrect agent group name

Fix: Ensure exact match with OIC configuration

2. Agent Shows Inactive

Cause: Network/firewall issue

Fix: Check outbound connectivity to OIC

3. Uneven Load Distribution

Cause: One agent slower than others

Fix: Check CPU/memory usage on server

4. SSL Handshake Errors

Cause: Certificate mismatch

Fix: Re-import certificates properly

Best Practices (From Real Implementations)

1. Always Use Minimum 2 Agents in Production

Single agent = Single point of failure

2. Deploy Agents on Different Servers

Avoid same machine deployment

3. Monitor Agent Health

Use logs and monitoring tools

4. Use Consistent Naming Convention

Example:

• AGENT_NODE_1
• AGENT_NODE_2
• AGENT_NODE_3

5. Perform Regular Failover Testing

Simulate downtime scenarios quarterly

6. Allocate Proper Resources

7. Avoid Overloading Single Agent

Scale horizontally instead of vertically

Summary

Oracle Integration Cloud Agent High Availability is not just a technical configuration—it is a business continuity requirement.

In real projects, I’ve seen integrations fail during critical operations simply because only one agent was deployed. With proper HA setup:

• Downtime is minimized
• Integration reliability improves
• Business processes remain uninterrupted

By deploying multiple agents under a single agent group, organizations can achieve:

• Load balancing
• Automatic failover
• Scalability

For deeper reference, always review the official Oracle documentation:

https://docs.oracle.com/en/cloud/saas/index.html

FAQs

1. How many agents are required for high availability in OIC?

Minimum 2 agents are required, but 3 agents are recommended for production environments to ensure better fault tolerance.

2. Does OIC automatically handle failover between agents?

Yes, OIC automatically routes requests to active agents within the same agent group without manual intervention.

3. Can agents be installed on the same server?

Technically yes, but not recommended. For true high availability, agents should be deployed on different servers or VMs.