ISUP Suspend and Resume: Keeping Voice Circuits Alive Under Faults
When an ISUP speech circuit crosses half a continent, the bearer path can fail in ways the call-control layer never predicted. A cable cut between Adelaide and Melbourne, a flooded pit in Far North Queensland, or a backhoe incident in suburban Perth can sever transmission while the signalling link that set the call up remains perfectly healthy. ISUP solves this asymmetry with two comparatively quiet messages: Suspend (SUS) and Resume (RES).
In Australian carrier networks, where the National Broadband Network and legacy PSTN switches still interconnect in regional Points of Interconnect, the gap between a working call and a dropped one often comes down to how those messages are handled. A correct Suspend and Resume exchange preserves the circuit identifier, the called-party number, and the charge records, so billing remains intact while technicians resolve the physical fault.
This article unpacks the message structure, the triggers that cause exchanges to send them, and the engineering practices used by operators from Telstra to smaller regional providers. The goal is practical: when a fault occurs, the switch behaves predictably and the customer hears a held line rather than a reorder tone.
Purpose of Suspend and Resume in ISUP
The fundamental reason SUS and RES exist is to decouple the user-plane (the 64 kbit/s timeslot carrying voice) from the control-plane (the SS7 messages that set the call up). A normal call assumes a continuous, usable circuit from origination to termination. When the transmission path degrades, that assumption breaks.
SUS is the way a switch tells the far end that it cannot carry audio on this circuit right now, but asks the peer not to tear the call down. RES is the matching announcement that the circuit is usable again. Together they preserve the call state across a temporary failure window that might last from a few hundred milliseconds to several minutes.
Australian operators must tune these behaviours carefully because of the country's geography. A fault on a Sydney–Perth transit route can take far longer to repair than one in a dense metro ring, so holding circuits open during extended outages is often preferable to releasing them and re-attempting.
Message Structure and Mandatory Parameters
SUS and RES are simple ISUP messages by design. SUS carries the affected circuit identification code, an optional user-to-user information field, and the parameter compatibility indicators expected on any ISUP message. RES is even leaner, adding only the indication that the suspended circuit is once more available for service.
A handy comparison for engineers reviewing traces is the parameter map for the two messages:
| Parameter | Suspend (SUS) | Resume (RES) |
|---|---|---|
| Message type | 0x0D | 0x0E |
| Circuit identification | Mandatory | Mandatory |
| Call reference | Optional | Optional |
| User-to-user info | Optional | Not present |
| Cause indicator | Optional | Not present |
| Compatibility indicators | Required | Required |
Both messages travel over the same MTP3 links that handled the original IAM, which is why MTP route monitoring in the SS7 fault management guide matters as much for these scenarios as for the initial call setup.
When a Switch Sends SUS
The decision to send SUS rather than releasing the circuit is usually driven by a transmission fault detector on the bearer side. Common triggers include slip detection on the E1 framing, excessive bit error rate thresholds, or loss of the timeslot mapping in the underlying cross-connect equipment. When the threshold is breached, the exchange's call-processing logic suspends the circuit before higher-level timers would interpret the silence as a called-party clear.
In Australian practice, this often occurs during planned works in the Pilbara mining corridor, where operators coordinate with mining tenants before major shift changes. The voice services supporting autonomous haulage and control rooms must stay logically live even when backhaul fibre is briefly rerouted, and SUS is what bridges that gap.
When RES Is Generated and What It Restores
RES is the symmetric response. Once transmission is restored, the originating side sends RES to the far end so that the previously held call is reconnected without re-dialling. Billing timers, which were frozen at the moment of SUS, resume from the point of RES. This continuity is one of the reasons ISUP remains relevant in a softswitch world: it preserves session semantics that VoIP SIP re-INVITE flows have to simulate.
Engineers at carriers such as Optus and TPG have documented cases in regional NSW where RES prevented what would otherwise have been a manual reconnection for hundreds of subscribers during bushfire-related outages. The held line simply continued after repairs, with no customer action required.
Timer Tuning and Operator Policy
The duration for which a circuit may remain suspended is governed by the Suspend Timer (T34 in many vendor implementations) and the matching Resume Timer. If RES does not arrive before the timer expires, the call is released and resources are reclaimed. Operators tune these values based on expected mean time to repair, jurisdictional regulations, and the cost of holding trunk capacity idle.
The Australian Competition and Consumer Commission reporting framework, which tracks outage durations and customer-impact minutes, gives carriers a strong incentive to set conservative Resume Timers. Holding a circuit for too long in a single failure counts against service levels, but releasing too quickly can fragment long-distance calls in transit. The compromise is usually a value between 60 and 180 seconds, adjusted by region.
Interaction with SIGTRAN and IP Backhaul
Modern Australian carrier cores increasingly use SIGTRAN, where M3UA or SUA carries ISUP over IP. SUS and RES travel inside the same SCTP associations used for the original call, which means the IP layer's own monitoring (heartbeats, path failure detection) can interact with ISUP-level suspension. A flapping SCTP path may cause multiple rapid SUS and RES exchanges that confuse downstream billing systems if they are not de-duplicated.
For engineers building hybrid cores, SS7 Training provides reference configurations that show how to keep ISUP state machines and SIGTRAN retransmission buffers aligned so that a single physical fault results in a clean SUS followed by a single RES, rather than a storm of retransmissions.
Testing, Monitoring, and Field Practice
A practical way to verify SUS and RES behaviour is to inject controlled bearer faults using a test E1 or a software-defined timeslot simulator, then watch the MTP3 trace for the expected message pair. Operators in Brisbane and Adelaide routinely run these exercises during maintenance windows to confirm that their restoration scripts produce the right message rather than a release.
Continuous monitoring of SUS and RES ratios per trunk group also acts as an early-warning signal. A rising SUS rate on a particular route can predict an underlying transmission problem days before customers notice. Teams that chart this metric alongside the daycare dog desensitization methods used elsewhere in the organisation find that the discipline of staged exposure and measured response applies equally well to telecom network conditioning.
When a fault does occur, the practical takeaway is straightforward: SUS and RES together hold the circuit open for exactly as long as the operator's policy allows, then restore it cleanly. Australian networks that treat circuit continuity as an active engineering target, rather than a side effect of call release, keep their customer-impact metrics clean and their long-distance calls whole.
