We're accountable and we own it, even when the bug wasn't ours
On a Friday morning in New Zealand, a hidden flaw in third-party software severed internet and phone connections across the South Island and lower North Island, leaving thousands of One NZ and 2degrees customers suddenly cut off from the digital threads that hold modern life together. The outage, which hit Christchurch and Wellington hardest, was not a dramatic act of sabotage but something quieter and more unsettling — a reminder that the infrastructure underpinning daily commerce, communication, and community is only as strong as its least visible component. One NZ's chief executive did not deflect blame toward the vendor, choosing instead to absorb accountability and commit to doing better — a posture that speaks to the weight leaders carry when the systems they steward fail the people who depend on them.
- A software bug from a third-party vendor silently dismantled connectivity across two islands, leaving homes and businesses with no calls, no data, and no warning.
- Businesses couldn't open, AA centres turned customers away at the door, and an IT support firm in Wellington was fielding distress calls before the morning was half over.
- Customers like one Hutt Valley resident spent precious time troubleshooting their own equipment, only to discover the silence was coming from the network itself — and that no one had thought to tell them.
- One NZ activated satellite backup services as a stopgap, offering partial relief through messaging apps while engineers worked to locate and patch the underlying fault.
- By mid-morning the bug was fixed and service was returning, with the CEO publicly owning the failure and pledging to prevent a recurrence — though without detailing how.
Friday morning, internet and phone service went dark across the South Island and lower North Island of New Zealand. Customers of One NZ and 2degrees found themselves cut off — no calls, no data — with Christchurch and Wellington absorbing the worst of it. The cause was a bug embedded in software supplied by one of One NZ's technology partners, and it took hours to find and fix.
The human cost accumulated quickly. An IT support company in Wellington fielded fifteen calls before noon from businesses locked out of their systems. AA centres across the affected regions closed their doors entirely — without internet, they couldn't process transactions or access records. One Hutt Valley resident spent time troubleshooting his own equipment before realising the fault lay with the network, not his hardware. What stung as much as the outage itself was the silence: One NZ hadn't reached out to explain what was happening.
As a partial workaround, One NZ opened its satellite service to more affected customers, allowing at least basic messaging and data access through apps like WhatsApp. 2degrees, which relies on One NZ's infrastructure, directed its own customers to wait for restoration. Spark reported no disruption to its network.
When One NZ's chief executive Jason Paris issued a public statement, he didn't point to the third-party vendor as a shield. "We're accountable and we own it," he wrote, acknowledging that the outage, though geographically contained, fell well short of acceptable. The bug had been fixed, service was returning, and he committed to ensuring it wouldn't happen again — a promise made without yet specifying what changes would follow. For the businesses and residents who lost their morning to an invisible fault in the network, the episode was a sharp lesson in how completely modern life depends on infrastructure that most people never think about until it disappears.
Friday morning, the internet went dark across a swath of New Zealand. Customers of One NZ and 2degrees woke to find their connections severed—no calls going through, no data flowing. The outage blanketed the entire South Island and crept into the lower North Island, with Christchurch and Wellington bearing the heaviest load. By mid-morning, One NZ's outage map was a constellation of red dots marking the places where people could no longer work, call, or connect.
The culprit was a bug buried in software written by one of One NZ's technology partners. It took hours to locate and repair. One NZ's chief executive, Jason Paris, acknowledged the failure in a social media post, framing it as a problem that "took them a while to fix." He didn't hide behind the third-party vendor. "We're accountable and we own it," he wrote, recognizing that the source of the problem mattered less than the fact that his company's customers had lost service. "Although it was in isolated parts of the lower North Island and South Island, it's just not good enough."
The practical consequences rippled outward quickly. An IT support company in Wellington, IT NEAR U, fielded fifteen calls before noon from customers locked out of their systems. The manager, Craven Coetzee, described them as "very frustrated, they can't work." One resident, Carey Griffiths from the Hutt Valley, spent considerable time troubleshooting on his own end before realizing the problem wasn't his equipment—it was the network itself. What bothered him as much as the outage was the silence. One NZ hadn't reached out to tell him what was happening.
Automobile Association centres across the affected regions simply closed their doors. Without internet, they couldn't process transactions or access customer records. The AA posted on social media asking people to check with their local branch before visiting, apologizing for the disruption and asking for patience while the issue was resolved. It was a stark reminder of how thoroughly modern business depends on connectivity—even a roadside assistance organization couldn't function without it.
One NZ activated its satellite service as a partial workaround, opening access to more customers in the affected areas so they could at least send messages and access data through WhatsApp. It wasn't a full solution, but it was something. 2degrees, which shared the outage because it relies on One NZ's infrastructure, acknowledged the problem on its website and directed customers to wait for One NZ to restore service. Spark, the third major carrier, reported no impact to its network.
By the time Paris issued his statement, the bug had been fixed and service was being restored. He committed to ensuring it wouldn't happen again, though he didn't specify what changes One NZ would make. The company's teams had been working since the early hours to bring systems back online. For customers and businesses across the South Island and lower North Island, the morning had been lost to a problem they couldn't see or control—a reminder that infrastructure failures, however brief, can halt entire regions.
Notable Quotes
Although it was in isolated parts of the lower North Island and South Island, it's just not good enough.— Jason Paris, One NZ CEO
Customers were very frustrated, they can't work.— Craven Coetzee, IT NEAR U manager
The Hearth Conversation Another angle on the story
Why does a software bug in someone else's code become One NZ's responsibility to apologize for?
Because One NZ is the company customers pay for service. The vendor wrote the code, but One NZ chose that vendor and deployed it. When it breaks, customers don't call the vendor—they call One NZ.
Did One NZ know this bug existed before Friday morning?
There's no indication they did. If they had, they would have fixed it already. This was a latent problem that only surfaced when the bug triggered under real-world conditions.
Why did it take so long to fix if it was just a software bug?
Finding the source of an outage across a network that large is detective work. You have to isolate which system failed, then trace it back through layers of infrastructure. Once they identified it was in the partner's software, they had to wait for that vendor to patch it.
What frustrated people most—the outage itself, or how One NZ handled it?
Both, but differently. The outage was unavoidable once it happened. What people resented was the lack of direct communication. Carey Griffiths spent time fixing his own equipment because he didn't know it was a network-wide problem. One NZ could have sent alerts faster.
Why did the AA have to close?
They run on the same internet as everyone else. Without connectivity, they can't access customer accounts, process payments, or even look up service records. A business that depends entirely on digital systems becomes non-functional the moment the network fails.
Is this likely to happen again?
One NZ says they're working to prevent it, but they can't control their vendors' code quality. What they can do is build better monitoring and faster response systems, and maybe diversify their technology partners so a single vendor's failure doesn't take down the whole network.