Summary

Document the 7 reliability pattern components including circuit breaker, retry policy, error boundary, graceful degradation, fault tolerance manager, resource manager, and data consistency.

Parent Issue

Part of: [EPIC] docs: Address documentation gaps across all ecosystem systems (kcenon/common_system#325)

Background (Why)

monitoring_system includes a comprehensive reliability subsystem at include/kcenon/monitoring/reliability/ with 7 components. While the API reference lists these types, there is no usage guide explaining when and how to apply each pattern, configuration best practices, or composition patterns.

Source files:

• include/kcenon/monitoring/reliability/circuit_breaker.h — CLOSED/OPEN/HALF_OPEN state machine
• include/kcenon/monitoring/reliability/retry_policy.h — Configurable retry strategies
• include/kcenon/monitoring/reliability/error_boundary.h — Error isolation
• include/kcenon/monitoring/reliability/graceful_degradation.h — Feature degradation under load
• include/kcenon/monitoring/reliability/fault_tolerance_manager.h — Orchestrates all patterns
• include/kcenon/monitoring/reliability/resource_manager.h — Resource limit enforcement
• include/kcenon/monitoring/reliability/data_consistency.h — Consistency guarantees

Scope (What)

Create docs/guides/RELIABILITY_PATTERNS.md covering:

1. Pattern Overview

• When to use each pattern
• How patterns compose together
• Fault tolerance manager orchestration

2. Circuit Breaker

• State machine: CLOSED → OPEN → HALF_OPEN → CLOSED
• Configuration: failure threshold, timeout, half-open attempts
• Monitoring circuit breaker state
• Example: Protecting exporter connections

3. Retry Policy

• Strategies: immediate, fixed delay, exponential backoff, jitter
• Max attempts and timeout configuration
• Idempotency considerations
• Example: Retrying failed metric exports

4. Error Boundary

• Error isolation between components
• Fallback behavior configuration
• Error propagation vs containment
• Example: Isolating collector failures

5. Graceful Degradation

• Load detection and degradation triggers
• Feature priority levels
• Recovery when load decreases
• Example: Reducing collection frequency under high load

6. Resource Manager

• CPU, memory, disk, network limits
• Enforcement strategies
• Alert integration for resource exhaustion

7. Data Consistency

• Consistency levels and guarantees
• Conflict resolution strategies
• Eventual consistency patterns

8. Composition Example

auto monitor = performance_monitor::create()
    .with_circuit_breaker(breaker_config{...})
    .with_retry(retry_config{...})
    .with_degradation(degradation_config{...})
    .build();

Acceptance Criteria

• All 7 reliability components documented
• Configuration examples for each pattern
• Composition patterns explained
• At least 4 real-world usage examples
• Decision guide: which pattern for which scenario