✨ Add AFTN protocol validation and serial reader health monitoring to enhance aviation telegram processing reliability and observability
Implement comprehensive AFTN/ICAO protocol compliance validation with configurable enforcement, enabling early detection of malformed telegrams and reducing downstream processing errors. Add real-time serial reader health monitoring to automatically detect message flow interruptions and sequence gaps, ensuring operational visibility into the telegram ingestion pipeline. Key enhancements: - AFTN validator validates priority indicators (FF/GG/QU/DD/SS/KK), ICAO addresses (4-char alphanumeric), and datetime formats (DDHHMM) with detailed error categorization - Invalid telegrams automatically routed to DLQ with full context for offline review and correction - Serial reader health monitoring tracks message gaps and sequence numbers to detect stalled readers or missing messages within configurable threshold (default: 2 minutes) - Four new Prometheus metrics expose validation errors by type, message gaps, sequence gaps, and health status for operational alerting - Pre-configured Prometheus alert rules for critical conditions (stalled reader, high error rates, consumer lag) - Grafana dashboard provides real-time visibility into AFTN compliance and serial reader health - Validation disabled by default for safe rollout with zero breaking changes to existing functionality Implementation maintains clean architecture with validator in adapter layer, extends processor and consumer with health tracking, and ensures thread-safe concurrent access to tracking state. All changes fully tested with 48 validator tests, 10 processor tests, and 21 consumer tests passing. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Sonnet 4.5
parent
6eff35b56f
commit
7b6f6383ad
@@ -22,6 +22,7 @@ type Config struct {
|
||||
Telemetry TelemetryConfig `koanf:"telemetry"`
|
||||
Monitoring MonitoringConfig `koanf:"monitoring"`
|
||||
DLQ DLQConfig `koanf:"dlq"`
|
||||
AFTN AFTNConfig `koanf:"aftn"`
|
||||
// Legacy fields for backward compatibility during migration
|
||||
Subscription SubscriptionConfig `koanf:"subscription"`
|
||||
Timeouts TimeoutsConfig `koanf:"timeouts"`
|
||||
@@ -146,6 +147,19 @@ type DLQConfig struct {
|
||||
Subject string `koanf:"subject"`
|
||||
}
|
||||
|
||||
// AFTNConfig defines AFTN protocol validation and monitoring settings
|
||||
type AFTNConfig struct {
|
||||
// ValidationEnabled enables AFTN protocol validation
|
||||
ValidationEnabled bool `koanf:"validation_enabled"`
|
||||
|
||||
// MessageGapThreshold is the duration after which the serial reader
|
||||
// is considered stalled (no messages received). Default: 2 minutes.
|
||||
MessageGapThreshold time.Duration `koanf:"message_gap_threshold"`
|
||||
|
||||
// EnableSequenceGapDetection enables monitoring for missing sequence numbers
|
||||
EnableSequenceGapDetection bool `koanf:"enable_sequence_gap_detection"`
|
||||
}
|
||||
|
||||
// MonitoringConfig controls the lightweight HTTP server that exposes health and metrics endpoints.
|
||||
type MonitoringConfig struct {
|
||||
Disabled bool `koanf:"disabled"`
|
||||
@@ -317,6 +331,11 @@ func LoadConfig() (*Config, error) {
|
||||
cfg.Monitoring.HealthTimeout = 2 * time.Second
|
||||
}
|
||||
|
||||
// Set AFTN defaults
|
||||
if cfg.AFTN.MessageGapThreshold == 0 {
|
||||
cfg.AFTN.MessageGapThreshold = 2 * time.Minute
|
||||
}
|
||||
|
||||
// Validate configuration
|
||||
if err := cfg.Validate(); err != nil {
|
||||
return nil, fmt.Errorf("config validation failed: %w", err)
|
||||
@@ -406,6 +425,10 @@ func (c *Config) Validate() error {
|
||||
if c.Monitoring.HealthTimeout < 0 {
|
||||
return fmt.Errorf("monitoring.health_timeout must be >= 0")
|
||||
}
|
||||
// Validate AFTN configuration
|
||||
if c.AFTN.MessageGapThreshold < 0 {
|
||||
return fmt.Errorf("aftn.message_gap_threshold must be >= 0")
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user