Skip to main content
Welcome back. In this lesson we cover AWS Systems Manager OpsCenter — the native incident tracking solution that centralizes operational and security incidents, structures them as OpsItems, and enables automated remediation with runbooks. Use this guide to understand OpsItem structure, event sources, runbook automation, cross-account aggregation patterns, and best practices for production operations.
A presentation slide titled "OpsCenter and Incident Tracking." It shows four colored boxes describing steps: Aggregate Incidents (collect from multiple AWS services), Structure Data (create OpsItems with context), Automate Response (execute runbooks for remediation), and Track Resolution (manage incident lifecycle).

What OpsCenter provides (at a glance)

OpsCenter centralizes incident records and offers:
  • Aggregation of incidents from multiple AWS services.
  • Structured incident records (OpsItems) with metadata and operational data.
  • Automated remediation by associating runbooks (Systems Manager Automation documents).
  • Lifecycle tracking from detection through remediation and resolution.
Use OpsCenter to reduce mean time to detection and resolution by combining automated event ingestion with runbook-driven remediation and audit trails.

OpsItems — structured incident records

OpsCenter stores incidents as OpsItems. Each OpsItem is a structured record containing metadata, operational data, and references to related AWS resources or external findings (for example, Security Hub or AWS Config results). This structure makes triage, automation, and auditing simpler and more consistent. Key parts of an OpsItem:
A presentation slide titled "OpsItems – Structured Incident Records" with three blocks labeled Core Metadata, Operational Data (highlighted), and Related Resources. The right panel lists operational data keys such as /aws/automations, /aws/resources, /aws/dedup and mentions custom organization-specific keys.
Related resources are the actual AWS resources connected to an OpsItem (for example EC2 instances, RDS databases, S3 buckets, Lambda functions, or CloudFormation stacks). Each OpsItem may reference up to 100 resources to give context to the incident and to support automated remediation.
A presentation slide titled "OpsItems – Structured Incident Records" showing three labeled sections (Core Metadata, Operational Data, Related Resources). The right panel lists related resources and notes (EC2, RDS, S3, Lambda, CloudFormation, EventBridge linking, manual addition via console/API, max 100 resources).

Event sources that create OpsItems

OpsItems can be created by a variety of event sources. Common integrations include: EventBridge is frequently used as the centralized ingestion point because it can translate and route events into OpsItem creation consistently and immediately.
A slide titled "Event Sources Creating OpsItems" showing a dotted box with icons and labels for CloudWatch Alarms, AWS Config, AWS Security Hub, AWS EventBridge, and Manual Creation. The image appears to outline different AWS event sources that can create OpsItems.
EventBridge is the recommended integration pattern for automated OpsItem creation because it enforces consistent payloads and reduces manual steps. Use EventBridge rules to transform service events into OpsItems so you can reliably trigger runbooks and downstream integrations.

Automatic vs manual OpsItem creation

  • Automatic creation: Preferred for high-frequency, machine-detectable incidents (CloudWatch alarms, AWS Config violations, Security Hub findings). Automation ensures consistent metadata and rapid response.
  • Manual creation: Useful for ad-hoc issues, external incidents, or findings from systems that cannot be captured automatically. Reserve manual creation for infrequent or exceptional cases.
Automate high-volume alerts and leave manual OpsItem creation for human-reported incidents to avoid noisy ticket practices and ensure scale.
A presentation slide titled "Automatic vs Manual OpsItem Creation" showing a boxed section about "Automatic Creation via EventBridge" with bullets explaining EventBridge auto-creates OpsItems (including AWS Config violations). To the right is a list of advantages like immediate incident creation, consistent metadata, no manual intervention, and reduced mean time to detection.

Example: EventBridge payload that creates an OpsItem

Below is an example target payload for an EventBridge rule that invokes Systems Manager to create an OpsItem. It includes a role and OpsItem parameters such as title, source, severity, category, and operational data pointing to affected resources:
This pattern lets you capture the important context (resource IDs, rule names) in the OpsItem so runbooks and human responders have the necessary information immediately.

Runbooks and automated remediation

Runbooks in Systems Manager Automation are the primary mechanism for automated remediation. You can associate one or more runbooks with an OpsItem and execute them automatically (triggered by EventBridge or Lambda) or manually from the OpsCenter console. Systems Manager records execution history alongside the OpsItem for auditing and post-incident review.
The image is a presentation slide titled "Runbooks and Automated Remediation" showing a three-step workflow: 01 Associate Runbooks, 02 Execute Remediation, and 03 Track Resolution, each with colorful icons and brief explanatory text. It includes a copyright note for KodeKloud.
Common runbook associations (examples): Use parameterized runbooks so the same Automation document can operate across different resource types by receiving resource IDs and remediation parameters from the OpsItem OperationalData.
A presentation slide titled "Runbooks and Automated Remediation" showing a boxed list of common runbook associations. Entries include high CPU utilization → AWS-RestartEC2Instance; disk space low → custom disk expansion; security group violation → custom security remediation; and patch non-compliance → AWS-RunPatchBaseline.

Third-party integration and cross-account aggregation

OpsItems can be integrated with ITSM platforms (ServiceNow, Jira) and aggregated from multiple AWS accounts. Integration patterns:
  • OpsCenter → EventBridge → Lambda → ServiceNow REST API
  • OpsCenter → EventBridge → API Gateway webhook → Third-party endpoint
For multi-account setups, use EventBridge cross-account event buses to forward events from source accounts into a central account that hosts OpsCenter. This central view simplifies runbook management and operations.
A slide titled "Cross-Account Incident Aggregation" showing a three-step vertical flow. The steps are in colored rounded bars labeled "1 Source Accounts", "2 Cross-Account Delivery", and "3 Central OpsCenter", connected by dashed arrows.
A presentation slide titled "Cross-Account Incident Aggregation" showing a three-step flow (Source Accounts → Cross-Account Delivery → Central OpsCenter) and a "Service Limit" note. The note lists "300 rules per event bus (soft limit)" and advises planning architecture carefully when aggregating from many source accounts.
When aggregating events across accounts, plan for EventBridge service limits (for example, 300 rules per event bus soft limit). Consider partitioning across multiple event buses, consolidating rules, or using filtering logic to avoid hitting limits in large organizations.

OpsCenter vs CloudWatch vs X‑Ray — complementary tools

These services each serve a distinct role in observability and incident response:
  • OpsCenter: centralized incident tracking and lifecycle management; correlates events, stores OpsItems, and integrates with ITSM tools.
  • CloudWatch: metrics, logs, dashboards, and alarms for detection and alerting.
  • X‑Ray: distributed tracing and latency analysis for root-cause analysis.
Common integrated workflow:
  • CloudWatch alarm detects a metric anomaly → EventBridge creates an OpsItem linked to CloudWatch dashboards/logs.
  • X‑Ray identifies a slow trace → OpsItem created and linked to the X‑Ray trace for troubleshooting. Combining these links provides end-to-end observability across metrics, logs, traces, and incident records.
A presentation slide titled "OpsCenter vs CloudWatch vs X‑Ray" showing three colorful panels with icons. Each panel briefly summarizes a service: OpsCenter (incident tracking), CloudWatch (metrics and logs), and X‑Ray (distributed tracing and latency analysis).
A flowchart titled "Integration Pattern for Comprehensive Observability" showing two paths: CloudWatch alarms automatically create OpsCenter items and link to CloudWatch dashboards, while X‑Ray identifies slow responses that lead to manually created OpsCenter items linked to X‑Ray traces. Both paths converge to "Complete Observability."

Best practices for production OpsCenter

  • Incident classification: Define severity tiers (Critical, High, Medium, Low, Informational) and standardize how alerts map to these tiers.
  • Automation and integration: Automate high-frequency incidents (CloudWatch, AWS Config, Security Hub) to reduce manual handling and MTTR.
  • Operational excellence: Track resolution metrics by category and severity with CloudWatch dashboards; iterate on runbooks and processes.
  • Data hygiene and retention: Archive resolved OpsItems older than your retention window (for example, >90 days) to S3 for compliance and cost control.
  • Auditing: Enable CloudTrail for all accounts to retain API activity and execution history for OpsItems and Automation runs.
A presentation slide titled "Production OpsCenter – Best Practices" showing four numbered items. It lists Incident Classification, Automation and Integration, Operational Excellence, and Data Hygiene with brief descriptions (define severity levels, automate CloudWatch/Config/Security Hub incidents, monitor resolution times with CloudWatch, and archive resolved items to S3).

Takeaways

  • AWS Systems Manager OpsCenter centralizes incident management and tracks incident lifecycles using OpsItems.
  • EventBridge is the standard integration pattern for automatic OpsItem creation from CloudWatch, AWS Config, and Security Hub.
  • OpsItems combine metadata, operational data, related resources, and runbook links to enable triage and automated remediation.
  • Use OpsCenter for incident workflow and tracking, CloudWatch for metric/log detection, and X‑Ray for distributed tracing; integrate them for full observability.
  • OpsItem storage is included, but automation and frequent API operations can generate API request costs — monitor usage in high-volume environments.
Further reading and references: That’s it for OpsCenter.

Watch Video