운영 장애 정의

Welcome to Incident Response Fundamentals! Our first step is to clearly define what a production incident is. It's not just any bug!

In the world of software, a bug is a known flaw or error in the code. An incident, however, is a sudden, unexpected event that disrupts normal service.

A production incident is an unplanned interruption to a service or a reduction in the quality of a service. It's something that prevents your users or internal systems from working as expected.

• It's unexpected.
• It causes negative impact.
• It requires immediate attention to restore normal operations.

While a bug might cause an incident, they aren't the same thing.

• A bug is a defect in the code (e.g., a button doesn't work).
• An incident is the impact of that bug or another issue (e.g., users can't complete checkout due to a broken button).

Incidents are about the *service disruption*, not just the underlying flaw.

The core of an incident is its impact. This can manifest in many ways:

• User-Facing Impact: Users can't access your website, app features are broken, or performance is very slow.
• Data Impact: Data loss, corruption, or unauthorized access.
• Internal System Impact: Critical backend services are down, preventing operations.

The wider the impact, the more severe the incident.

Not all incidents are equally critical. We use severity levels to categorize incidents based on their impact and urgency. This helps teams prioritize their response.

Commonly, incidents are rated from S1 (most severe) to S4 (least severe). Let's dive into what each level means.

An S1 incident is the highest level of severity. These are catastrophic issues that demand immediate, all-hands-on-deck attention.

• Characteristics: Complete service outage, major data loss, significant financial impact, widespread user impact.
• Example: Your main website is entirely down, and no users can access it globally.

An S2 incident indicates a significant degradation of service or a partial outage. It affects a large number of users or a critical business function.

• Characteristics: Major feature is unavailable, widespread performance issues, partial data loss, significant customer dissatisfaction.
• Example: Users can log in, but the payment processing system is completely failing, preventing purchases.

An S3 incident represents a minor service degradation or a non-critical feature being unavailable. Workarounds often exist, and the impact is limited.

• Characteristics: Small subset of users affected, non-critical feature broken, minor data discrepancy, performance slightly degraded.
• Example: The 'contact us' form on your website is broken, but users can still email support directly.

An S4 incident is the lowest level of severity. These are typically cosmetic issues, minor bugs, or very low-impact problems that do not significantly affect service functionality or user experience.

• Characteristics: Typo on a static page, minor UI glitch, very limited internal impact.
• Example: A deprecated link appears on a rarely visited internal dashboard.

Great job! In this lesson, we've learned the fundamental definition of a production incident: an unplanned service disruption with negative impact.

• We distinguished incidents from everyday bugs.
• We explored different types of impact, from user-facing to internal.
• Most importantly, we introduced and defined the common severity levels (S1-S4) to help prioritize response.

Understanding these concepts is crucial for effective incident response!