smpp-connector — Observability

Status: populated | Last updated: 2026-04-18

1. Metrics (Prometheus)

All metrics exposed at GET /metrics on port 3001.

Metric name	Type	Labels	Description
smpp_connector_sessions_total	Gauge	operator_id, state	Current sessions by state (BOUND/UNBOUND/CONNECTING)
smpp_connector_submit_sm_total	Counter	operator_id, status (success, error, throttled)	submit_sm PDUs sent
smpp_connector_submit_sm_duration_seconds	Histogram	operator_id	Time from submit_sm send to submit_sm_resp
smpp_connector_dlr_received_total	Counter	operator_id, status	deliver_sm PDUs received by DLR status
smpp_connector_dlr_correlation_not_found_total	Counter	operator_id	DLRs received with no matching correlation record
smpp_connector_tps_throttle_total	Counter	operator_id	Number of messages throttled by TPS limiter
smpp_connector_tps_current	Gauge	operator_id	Current TPS rate (last 1 s window)
smpp_connector_reconnect_attempts_total	Counter	operator_id	Total reconnection attempts
smpp_connector_enquire_link_timeout_total	Counter	operator_id	enquire_link timeouts (precede UNBOUND transitions)
smpp_connector_pdu_bytes_sent_total	Counter	operator_id	Total bytes sent over SMPP socket
smpp_connector_pdu_bytes_received_total	Counter	operator_id	Total bytes received over SMPP socket
smpp_connector_long_message_segments_total	Counter	operator_id, strategy (csms, tlv)	Segmented PDUs sent

Key Alerts

Alert	Condition	Severity
SmppConnectorSessionUnbound	sessions_total{state="BOUND"} == 0 for any operator for 60 s	critical
SmppConnectorHighThrottleRate	rate(tps_throttle_total[1m]) > 10 per operator	warning
SmppConnectorHighDlrCorrelationMiss	rate(dlr_correlation_not_found_total[5m]) > 1	warning
SmppConnectorEnquireLinkFailure	enquire_link_timeout_total increasing for 2 min	warning
SmppConnectorAllSessionsDown	All operator sessions UNBOUND simultaneously	critical

---

2. Structured Logging

All logs emitted as JSON to stdout.

Standard log fields

Field	Type	Description
timestamp	ISO 8601	Event time
level	string	debug / info / warn / error
service	string	Always smpp-connector
operatorId	string	Operator context for the log event
traceId	string	OpenTelemetry trace ID (propagated from NATS message headers)
message	string	Human-readable summary

Key log events

Event (message)	Level	Extra fields
smpp.session.connecting	info	operatorId, host, port
smpp.session.bound	info	operatorId, bindMode, reconnectAttempts
smpp.session.unbound	warn	operatorId, reason
smpp.session.failback	info	operatorId
smpp.submit_sm.sent	debug	operatorId, sequenceNumber (no to, no text)
smpp.submit_sm.resp	debug	operatorId, sequenceNumber, operatorMessageId, commandStatus
smpp.submit_sm.throttled	warn	operatorId, tpsCurrent, tpsLimit
smpp.dlr.received	info	operatorId, operatorMessageId, messageState
smpp.dlr.correlation.not_found	warn	operatorId, operatorMessageId
smpp.enquire_link.timeout	warn	operatorId
smpp.reconnect.scheduled	info	operatorId, attempt, delayMs

PII policy: to (destination number) and text (message content) are never logged.

---

3. Distributed Tracing (OpenTelemetry)

• Trace context propagated from NATS message headers (traceparent header per W3C Trace Context).
• Spans created:
  • smpp-connector.dispatch (per NATS message consumed)
  • smpp-connector.tps_check (Redis TPS window check)
  • smpp-connector.submit_sm (PDU send + response wait)
  • smpp-connector.db.insert_correlation
  • smpp-connector.dlr.process
  • smpp-connector.dlr.publish