On the 18th of September 2025, Optus – an Australian telecommunications company – experienced a strange network outage affecting triple-zero calls, but not standard and international calling. While the outage has been resolved, many families have been left in disarray with their cries for help unheard, including three deaths that can be attributed to it. Optus is now undergoing senate hearings with the Federal Senate to create accountability for these actions and mandate plans to prevent deadly outages like this from occurring again.
The first, of many, senate hearings occurred on Monday 8 AM AEDT, November 3. This hearing was held to gain a holistic overview of the events that had occurred and how Optus had responded. It gave the Senate a chance to apply pressure on Optus given very little communication was sent to the Minister for Communications, The Hon. Anika Wells and/or the Australian people.
The three and a half hour long proceeding began with and opening statement from Optus, delivered by Mr Stephen Rue – the CEO of Optus, former CEO of NBN Co. Mr Rue explained that the outage experienced in the evening of the 18th of September 2025 was due to a bad firewall policy upgrade, causing triple-zero calls to not be connected while standard and international calls could be placed as normal. This issue was compounded by the fact that the technicians selected the wrong process plan while performing the upgrade and warnings and alarms were raised but misinterpreted thus not acted on.
To improve from this deadly upgrade, the network have made some changes (but not limited to);
- Explicit confirmation that triple-zero calls are operational before and after any upgrades
- 24/7, state-by-state monitoring of the triple-zero system
- Contact centre procedures are being overhauld
- Any reports of a triple-zero outage should be escalated immediately to the networks team for investigation
- Introduced daily manual testing of the triple-zero service (i.e. someone calling 000 on their phone) – with plans to automate this.
- Start moving call centres back on shore and employ another 300 people (in particular to the special care team – they handle regional, elderly and disabled customers)
Mr Rue joined Optus a year ago “because I know it’s important for Australia that it [Optus] succeeds” ~ Opening Statement, Senate Hearing Mon 3rd November, Mr Stephen Rue. He would be making company-wide transformations to create process improvements, clearer accountability and capability uplift. Mr Rue regrets “our reforms were not embedded sooner to prevent the events of September 18” and he firmly believes that a change of leader at this time is not what Optus needs. It could set back a multi-year plan to improve the company, that we are already a year into.
In the late evening of the 18th, the call centre staff were making usual welfare calls to customers that attempted to dial 000, but were not connected. Since a customer can naturally lose connection to the network (i.e. walking into a dead-spot), this was standard practice for Optus. However as the night went on, the list of welfare calls to be completed was increasing. The team began to suspect there was an outage, which was confirmed soon after. The Optus networks team discovered that the CAMPON service – the protocol that allows a customer of Optus to use another carrier for emergency dialing – was dropping all incoming connections, the service had become locked down (in layman’s speak, shutdown). The call centre and networks team begun finding other, unusual, ways to access the list of numbers that CAMPON had dropped. The list of welfare calls quickly mounted to over 100.
By half past midnight, the Optus Community Manager emailed management informing of two deaths that had occurred a few hours prior. This email will go unread until approximately 8 AM the following morning.
By the morning of September 19, the outage had already resolved. Optus focused their efforts on data gathering and accuracy checking to package everything together for the Australian Government and the regulator – the Australian Communication and Media Authority (ACMA here forth).
It is standard corporate practice for a CEO to inform the board of directors and a member on the board of a group company (if there is one) of large circumstances like this. Usually, Mr Rue would contact the Mr John Arthur – Chairperson of Optus, who would then pass the message onto the group company, Singtel. At the time Mr Arthur was overseas (about 7 hours behind AEDT), a call was attempted but went to voicemail. Mr Rue then contacted Mr Yuan Kuan Moon – CEO of Singtel – to inform Singtel of the outage and “killing two birds with one stone” ~ Rue, as both the CEO and board needed to be alerted of the outage.
From this initial senate hearing it was unclear exactly what took place on the 19th. Between the hours of 8 AM (when management heard about the outage), 2:30 PM (when the regulator was informed) and 4 PM (when the Minister’s office was informed) the government was “kept in the dark”. Mr Rue claims that welfare checks were still underway in the morning and the board and himself were in 11 directors’ meetings throughout the day to organise the company and compile as much accurate documentation as possible.
Although there was no immediate danger to the Australian people on the Friday, many still believe Optus handled the morning incorrectly. This is one of the major reasons as to why the senate was formed. The last update from Optus to the government was from the 18th, when the network was operational. It seems given the outage occurred late at night, the fact the Minister’s office wasn’t informed until the next day was lessened. However, the board discovered of the outage and the fatalities as of 8 AM on the Friday and “information was left sitting devastatingly wrong, deadly wrong, for six hours” ~ Chair of the Senate, Senator Handson-Young.
The documentation gathered by Optus describes that the outage was caused by a bad firewall policy change, which was not peer-reviewed or tested. The technician had selected the wrong processes and did not follow the updated procedure. The only documentation submitted was a risk management plan. After the upgrade was pushed, alerts of a triple-zero fault were misinterpreted and not taken as errors from the upgrade. Call centre staff were also not acting correctly on customers’ reports of unconnected triple-zero calls.
Although these technical procedures were not required, the ACMA had introduced new mandates and regulations in April of 2025 (to be enforced on the 1st of November) – two years after the Bean Review (view the outcome here). The telco was “sitting on its hands for two years” ~ Senate Chair.
Controls for situations similar to this were in place. The processes worked, in fact this one was a good process (according to Mr Rue), but the problem was the individuals did not follow them. A “catastrophic mistake [was] made by Optus” ~ Senator Sarah Henderson, Senator for Victoria, and we hear too often that these types of outages are down to human error. Granted, humans make mistakes, but we should all know that (we all have made mistakes). Redundancies and backup plans should be created during the design phase of any project and regularly tested once deployed. Especially for a life-saving system like CAMPON and the triple-zero network.