November 18, 2026

OIC Monitoring: What to Watch Everyday in Oracle Integration 3

Key Takeaways

  • Oracle patches and upgrades Oracle Integration 3 for you, but its responsibility matrix assigns integration monitoring to the customer.
  • Oracle Integration 3 includes Observability pages for dashboards, instances, errors, scheduled runs, connectivity agents, and Oracle Fusion event subscriptions.
  • Oracle recommends daily checks for errors, timeouts, performance, recent changes, falling instance counts, agent availability, and the connections with the most errors.
  • Service limits shape the routine. Failed messages aren’t resubmitted automatically, and production tracking data is kept for 32 days by default.
  • The week after an Oracle Fusion quarterly update deserves a closer watch, starting with a comparison against your normal instance counts and error rates.

Every Oracle Fusion team that runs Oracle Integration 3 shares the same arrangement with Oracle. Oracle “completes functional and security patching updates work on your behalf, with no work required by you.” The integrations themselves are a different matter. In the same responsibility matrix, Oracle lists the customer as responsible and accountable for integration monitoring and management, and advises: “Make sure you’re also monitoring events generated outside of Oracle Integration.”

That makes OIC monitoring part of operations, not a task for when something breaks. This guide covers what Oracle Integration 3 gives you to work with, a daily routine based on Oracle’s own monitoring workflow, the service limits that shape it, and what to watch after an Oracle Fusion quarterly update.

Who Owns What in Oracle Integration 3

Area Oracle Your team
Platform patching and upgrades Responsible and accountable Informed
Connectivity agent upgrades Responsible and accountable Informed
Integration monitoring and management Consulted Responsible and accountable
Health of on-premises connectivity agents Consulted Responsible and accountable
Promotion from test to production Consulted Responsible and accountable
Source control and CI/CD Consulted Responsible and accountable

Source:Oracle Integration 3: Oracle and customer responsibilities

The connectivity agent row is easy to miss. Oracle notes that it “doesn’t have access to the servers that host your on-premises connectivity agent or the agent logs.” When an agent goes down, every integration that depends on it fails, and nobody at Oracle will see it first.

What Oracle Integration 3 Gives You to Monitor

Oracle Integration 3 groups its monitoring tools under Observability pages:

  • Dashboards give “at-a-glance information about the messages that Oracle Integration processed and the integrations that ran.”
  • Instances track every integration run and its message payload.
  • Errors list only the instances that failed.
  • Future runs show scheduled integrations and their timing, so you can avoid starting too many at once.
  • Agents identify which connectivity agent is down.
  • Design time audits show who changed an integration, and when.
  • Fusion Applications help find the source of failures in integrations that consume events published by Oracle Fusion Cloud Applications.

Outside the Oracle Integration console, the Oracle Cloud Infrastructure (OCI) console adds service metrics, custom charts, and alarms that “notify you when metrics meet specified triggers.” Teams that need deeper analysis or longer history can also send the activity stream to OCI Logging.

A Daily OIC Monitoring Routine

Oracle’s workflow for monitoring integrations lists the checks to run “at least daily, often at predetermined times.” Here they are, with where to look and why each one matters.

Daily check Where to look Why it matters
Did any integrations fail? Errors page Each failure is a transaction that didn’t reach its destination
Did any time out or get aborted? Instances page, filtered by status Timeouts often point to a slow or unavailable endpoint
Are integrations running slower than usual? Dashboards, Integrations page Slow runs can push scheduled work past business deadlines
Did anyone change an integration recently? Design time audit Oracle notes that “an inadvertent update could be to blame” when something stops working
Are fewer instances running than normal? Dashboards, Instances page A drop can mean an upstream system stopped sending data
Are all connectivity agents available? Agents page If an agent is down, every integration that uses it fails
Which connections have the most errors? Errors page, grouped by connection Shows where to start when several things fail at once

Oracle also recommends a few occasional checks: confirm that scheduled integrations still have staggered start times, keep active integrations below your limits, and avoid the Debug tracing level in production.

Service Limits That Shape OIC Monitoring

Two limits decide how fast your team needs to act, and how long you can look back.

Failed messages wait for you. Oracle’s documentation states that “Oracle Integration does not automatically resubmit failed messages.” Failed instances of asynchronous flows can be resubmitted manually, up to ten times and within a recovery window. Synchronous flows, and scheduled flows with scheduled parameters, can’t be resubmitted, and a discarded message can’t be recovered.

Tracking data has an expiry date. By default, Standard and Enterprise edition instances retain data for 32 days, and Healthcare edition instances for 184 days. Only Enterprise edition customers can change the period. Debug tracing data is kept for 24 hours, and audit tracing data for 8 days. If you need trends beyond a month, or evidence for an audit, plan to export it.

What to Watch After an Oracle Fusion Quarterly Update

Oracle Fusion updates arrive four times a year and can’t be skipped. Oracle describes its integration tools as backward compatible, and still advises customers to confirm they continue to work as expected. In the first week after an update reaches production, add these checks to the daily routine:

  1. Compare instance counts with a normal week. A drop in runs for an integration that pulls from Oracle Fusion is often the first sign that something changed upstream.
  2. Watch errors by connection. A spike on the Oracle Fusion connection points to the update. A spike elsewhere points to a different cause.
  3. Check event subscriptions. Use the Fusion Applications and Subscriptions pages for integrations that consume Oracle Fusion business events.
  4. Confirm scheduled runs completed. Check Future runs and the Instances page for jobs that should have run overnight.
  5. Compare running times. Oracle recommends comparing average running times after updates “to ensure that you haven’t introduced performance regressions.”
  6. Look beyond Oracle Integration. Confirm with the business that invoices loaded, approvals arrived, and files reached banks and payroll providers.

Keep a baseline from the week before the update. Without one, “fewer instances than normal” is hard to prove.

Let Problems Find You

Daily checks work best with alerts behind them.

  • Notifications: Oracle Integration can send “daily or hourly emails about failures or the overall health of all your integrations.”
  • OCI alarms: set them on the metrics that matter most, such as failed messages for your highest-impact flows.
  • A contact list: document each integration, the applications and connectivity agents it uses, and Oracle’s recommended “contact names with backups for who gets involved and informed in the event of a failure.”

Where IT Convergence Helps

IT Convergence’s Smart Integration Automation framework builds this routine into operations, with automated health checks, event monitoring, and root-cause analysis across Oracle Fusion, Oracle Integration, and the systems around them. It works alongside ITC’s integration services and Oracle Cloud Applications managed services, so failures get caught and acted on within the recovery window.

Frequently Asked Questions (FAQs)

  1. Does Oracle monitor my OIC integrations?
    Oracle monitors and patches the Oracle Integration platform. Its responsibility matrix assigns integration monitoring and management, and the health of on-premises connectivity agents, to the customer.
  2. How long does Oracle Integration 3 keep instance data?
    32 days by default for Standard and Enterprise editions, and 184 days for the Healthcare edition. Enterprise edition customers can change the period. Debug tracing data is kept for 24 hours and audit tracing data for 8 days.
  3. Does OIC retry failed messages automatically?
    No. Failed asynchronous instances can be resubmitted manually, up to ten times within a recovery window. Synchronous flows can’t be resubmitted.
  4. Can Oracle Integration 3 send alerts?
    Yes. Oracle Integration can email daily or hourly notifications about failures and overall health, and OCI alarms can notify you when metrics meet the thresholds you set.
  5. What should we monitor after an Oracle Fusion update?
    Instance counts against a normal week, errors by connection, Oracle Fusion event subscriptions, scheduled runs, running times, and the business results outside Oracle Integration, such as loaded invoices and delivered files.

See the Approach on Dec. 10

Join Achutha Shivashankar, Head of Cloud Practice at IT Convergence, for Oracle Fusion Updates: What Could Break and How to Prepare on Thursday, December 10, at 12:00 PM CT (1:00 PM ET). In 45 minutes, he’ll cover what to assess before an update, how to test connected processes, and what to watch after go-live, then take your questions live.

Save my seat

Debra Castillo
Debra Castillo Debra has over 11 years of experience working across Cloud Infrastructure, Managed Services, Workload Migrations, Tariff Impact Management, and Nearshoring. She specializes in writing compelling narratives to address complex challenges.

Related Blogs