On Wednesday morning, Telstra's mobile network collapsed due to a software defect that told the network it was back in November 2006. The glitch originated in the company's GPS node, which provides accurate time to the nanosecond for the entire network. Telstra CFO Michael Ackland explained that the reset changed the time and synchronisation, which then propagated to the rest of the network, effectively dialling back the internal clock to 2006.
How a time-telling error brought down the network
Smartphones automatically adjust for daylight savings using Network Time Protocol (NTP), which requests accurate time from dedicated servers. On Wednesday, the NTP within Telstra's GPS node suddenly broadcast a time nearly 20 years in the past. Networks rely on accurate time for synchronisation and authentication, Ackland said, "so that one part of the network knows what the other part is doing." Without correct time, messages or instructions from different time periods could be authenticated, causing security failures.
Khurram Shahzad, senior analyst at Gartner, explained that when the root clock fails, "the entire digital domino chain falls within minutes." Network nodes from cell towers authenticate their identity with the central data hub using security certificates with strict validity windows checked against the current time. If Telstra's central time servers broadcast the wrong time, core network switches see valid certificates as expired or invalid, assume a security threat, and terminate connections, locking devices out and throwing phones into "SOS only" mode. Shahzad noted that 4G and 5G also require precise time measures for optimal speeds.
Impact on emergency services
During the outage, some customers attempting to dial triple zero could not be connected, despite failsafe systems that should route emergency calls through other mobile networks. Telstra has not yet explained why this handover failed, but said it resolved the issue by Thursday. A second issue, also attributed to the time-keeping error, caused some customers to receive error messages when calling other numbers or triple zero. As of 1pm AEST on Thursday, Telstra had completed 639 welfare checks on people who made unsuccessful triple zero calls. Of those, 230 needed no assistance, 402 required a follow-up call, 170 were referred to police for welfare checks, and seven were referred to emergency services.
Disruption to rail services
The Australia Rail Track Corporation (ARTC) suspended services on Wednesday after the Telstra outage disrupted the national train communications system, which uses Telstra's 4G network for drivers to communicate with control centres. ARTC operates the Ghan, XPT, V/Line, and other services. V/Line CEO William Tieppo said Telstra's 4G network interfered with the satellite backup system used by train drivers. Overnight, Telstra repaired programming to stabilise the network, ensuring the 4G train radio system integrated correctly with satellite phones.
Telstra's response and network resilience
Ackland rejected comparisons to the Y2K bug, describing the incident as a software glitch. He said Telstra would conduct a root cause analysis with vendors, partners, and external support. Asked if the outage showed the network was fragile, Ackland replied, "It is a complex system. I think it is robust. There is a lot of redundancy, but this was an unfortunate incident." Telstra continues to work with a small number of enterprise customers on residual flow-on impacts.
