Detecting MTP2 link failures with SUERM and alignment monitoring
SS7 links can appear available while carrying corrupted signalling units. At MTP2, the signalling unit error rate monitor (SUERM) and alignment error rate monitoring provide complementary evidence: one measures received signalling-unit quality, while the other exposes loss of synchronisation with the bearer.
Reliable detection combines those measurements with MTP3 link management, spare-route capacity and controlled recovery. The aim is not merely to silence a noisy circuit, but to ensure that traffic moves to a healthy route without causing a second failure.
MTP2's role in link failure detection
MTP2, or Message Transfer Part level 2, provides the transport layer beneath MTP3. It assembles and validates signalling units, maintains frame alignment, reports link status and applies flow control. MTP3 uses that service for SS7 routing, while ISUP uses MTP3 for call-related signalling.
Failure detection therefore belongs below ordinary call-routing logic. A circuit can remain physically present yet fail its error checks, lose alignment, or become congested. Congestion is a separate condition: backpressure from a busy peer must not be misclassified as an unavailable link.
Reading SUERM counters
SUERM counts signalling units received during an observation window and tracks how many contain detected errors. The resulting error rate gives a quality signal even when alignment survives. Erroneous units can be rejected while correctly formed units continue to pass, allowing the monitor to reveal gradual degradation.
The important variables are the measurement interval, error threshold and recovery criteria. Compare errored units with the total observed, not with a fixed number of calls. After an alarm, require a healthy interval before returning the link to service; this hold-down prevents a flapping bearer from being selected repeatedly.
Alignment error rate monitoring
Alignment error rate monitoring examines whether the receiver can keep the incoming bit stream synchronised. If the alignment pattern cannot be found or sustained, the link enters a loss-of-alignment state and normal signalling units are not accepted. The monitor captures failures that a unit-error counter alone can obscure.
Persistent alignment loss usually requires immediate withdrawal from service. Intermittent errors need more diagnosis: they can arise from timing trouble, a faulty cross-connect, degraded wiring or an incompatible configuration. Distinguishing these causes avoids disabling a sound MTP2 peer when the real fault is in the bearer.
Converting measurements into link state
A practical detector combines the two indicators with timers, hysteresis and MTP3 route availability. A small rise in SUERM may justify investigation; a confirmed loss of alignment normally justifies immediate isolation. Vendor-defined thresholds and carrier acceptance criteria should determine the exact limits rather than an assumed universal percentage.
Before withdrawing a link, confirm that a spare MTP3 route has capacity and that the failed bearer is not a basic link supporting several traffic links. Record the state change, trigger the appropriate link-management procedure and observe recovery traffic before restoring the original route.
MTP3 and ISUP during rerouting
Once MTP2 reports a link failure, MTP3 can test a surviving path with the Signalling Link Test Message (SLTM) and its acknowledgement, SLTA. It then reroutes affected traffic according to its routing procedures. These timers and acknowledgements confirm signalling reachability; they do not themselves measure bearer error rate.
ISUP-level changes must be checked separately. The guide to call diversion indicators shows how upstream decisions can alter call routing; engineers must confirm whether a change began above MTP3 rather than in the underlying transport. A released call is not evidence that the failed SS7 link recovered.
Australian network considerations
Australian carriers still combine legacy PSTN and ISUP interconnects with IP-based fixed and mobile networks. Sydney and Melbourne are major interconnection centres where a bearer may serve several SS7 neighbours. Elsewhere, SIGTRAN adaptations may carry SS7 over IP, so transport alarms and MTP2 alignment indicators must be mapped to the actual architecture.
Long-distance routes to Brisbane, Perth, Adelaide and Canberra can involve leased E1 capacity, terrestrial fibre paths and remote sites. ACMA-regulated numbering and number portability also make routing changes operationally significant. Australian field-maintenance customs, including negotiated access windows and technician travel, make remote diagnosis and controlled link swaps particularly valuable.
Testing and maintaining detection thresholds
Validate thresholds during a controlled maintenance window rather than waiting for a production outage. Record baseline SUERM and alignment counters during a busy period and during a quiet one, then compare them with SLTM results and carrier fault reports. A test tool can introduce controlled signalling errors, but only where the network owner has approved the procedure.
Track each threshold change, timer and recovery action in the runbook, including the responsible exchange and access details. Keep the link out of service until spare-route traffic is stable, then restore it gradually. Monitoring should expose the counters to operations staff without weakening authentication or other signalling-security controls.
A useful operating model is to treat SUERM as a measure of signalling-unit quality and alignment monitoring as a test of framing. Use the first for degradation analysis and the second for rapid loss-of-synchronisation detection, with MTP3 responsible for confirming that an alternative route is usable.
The next concrete step is to select one production SS7 link, export its SUERM and alignment counters for a representative busy period, and compare them with SLTM results and switch events before approving any threshold change.
