Several New Zealand websites caught up in a global internet outage late last night are back up and running this morning.
The outage not only affected New Zealand but also servers in North America, India, Europe, Britain, Australia, Japan and South America although the failure was not geographically universal. Users in some locations, such as Berlin, reported no problems, while others experienced massive failures across the internet. Outages were reported in locations as varied as London, Texas and Australia.
Within minutes of the outage, a content delivery network company (CDN), a cloud computing services provider, acknowledged that its content distribution network was the cause of the problem. The company runs an “edge cloud”, which is designed to speed up loading times for websites, protect them from denial-of-service attacks, and help them deal with bursts of traffic. It is one of the biggest CDN providers in the world. While governments and service providers may still be unsure about what happened and the CDN company not specifically elaborating on the cause, the worldwide disruption is a harsh reality-check of the fragility that an interconnected internet infrastructure can be.
New Zealand was significantly hit with all sites down. Users trying to access the websites experienced their site labelling “503 errors connection failure”. However, with robust critical event management in place, the internet was back up and running again at 10.27 pm after crashing at 9.46 pm.
In an update at 9.58 pm NZT, the CDN firm said that it was investigating the issue. It provided several updates in the following 40 minutes, saying it was continuing to investigate the problem. At 10.44 pm NZT, it was reported that the issue had been identified and that a fix was being implemented but no reason given for the major outage.
Although internet disruptions are rare, they are unavoidable. No matter how competent your internet service provider is, unexpected circumstances can emerge and create an outage, potentially leaving people stranded in the middle of whatever online activity there were engaged in at the time.
The outage occurred when the CDN developed technical difficulties, disrupting national and international news sites, causing other domains to crash and being unavailable for up to an hour. CDNs are global networks of servers that work together to provide content over a large area and deliver it to users more rapidly, regardless of where they are in the world. Content can be cached to a CDN server near users so that it does not have to be fetched from the original server each time. The technology is thought to improve reliability by distributing website delivery over multiple sites rather than depending on a single data centre.
When a CDN malfunctions, an outage occurs, bringing all websites around the world to be offline. One of the main features of CDNs is that they have redundancy in place to prevent crashes like the one that had occurred. It is this that makes the outage a significant incident. It’s currently unknown whether the CDN outage was caused by a cyber-attack or a server breach.
Regardless, all internet service providers, big or small, are susceptible to internet outages. An internet outage usually means any issue that prohibits access to the internet. As a result, when an error message is displayed or the browser times out while loading a web page, the global network connection may not have necessarily failed. It is more likely because of a technical issue or could be local connection challenges. These can range in severity from large-scale failures to minor problems. Common causes when an outage occurs are when a server is down for maintenance or is overloaded.