MONDAY, 10.30am: Telecom has confirmed that only customers of its retail ISP, Telecom Broadband, were affected.
"The root cause of the issue is still under investigation," a spokeswoman said.
The outage lasted around one and a half hours.
SUNDAY, 4pm: New Zealand was partially cut off from the world this afternoon.
A widespread Telecom problem with international data, which struck shortly after 3pm, stopped New Zealanders accessing sites hosted overseas this afternoon - and services such as Gmail. Local sites with ads served from offshore services such as Google's Double Click have also been hit.
A Telecom spokesman had no immediate comment, but was seeking more information.
The outage appeared to only affect Telecom Broadband customers; Orcon confirmed none of its customers were affected.
The company has posted a blanket statement on its website in the meantime: "Some broadband customers might be experiencing issues browsing international websites. Our technicians are currently working on this problem."
The problem seemed to have been fixed for most users by 4pm - although a the service warning remained on Telecom's website.
Xero chief executive Rod Drury was one of the most vocal online during the outage.
At 3.56pm Mr Dury thanked Telecom International for fixing the problem, but added, "Important to have full explanation soonest. Outages [are] a major cost for cloud industry."
Xero's "software-as-a-service" online accounting serviced is hosted by a US company called Rackspace.
"Xero wasn't down, but New Zealand customers trying to access Xero and other global sites had to stop work," Mr Dury told NBR shortly after 4pm.
Mr Drury, along with fellow NBR Rich Listers Sam Morgan and Sir Stephen Tindall, is one of the founders of Pacific Fibre, which aims to break the 50% Telecom-owned Southern Cross Cable's monopoly on fibre running out of New Zealand.
As of 4.05pm, a Telecom spokesman had no official update, but offered that it seemed to be "an intermittent issue".
"These things happen, but it's very important to get a clear explanation of what did happen so that the risk of future outages can be minimized and those providing service's over the top can plan accordingly," Mr Dury said.
"This didn't appear to be a cable issue but shows how vital redundancy is for all layers in the stack.
"The real cost to us was dealing with customer [support query] tickets, lost productivity of our customers and any customers lost because of the ensuing FUD [fear, uncertainty and doubt.
"Even with this outage, cloud computing generally will have more availability than on premise solutions but glitches like this naturally rock confidence."
This article is tagged with the following keywords. Find out more about MyNBR Tags
Most listened to
- Fonterra Shareholders' Council chairman Duncan Coull says new study needed to restore confidence among shareholders
- Spoke Phone chief executive Jason Kerr explains what his app can offer
- Accountants give their first impressions of Labour's Tax Working Group
- Calida Smylie runs the rule over Air NZ's handling of the Dreamliner engine debacle
- NBR Radio: The best interviews – updated daily, with Grant Walker