Sanity News

8 of the Worst Network Failures and How You Can Avoid Them

Friday, September 22, 2017

Cyberattacks, human error, fires, natural disasters, and technical glitches can plague your servers at any time, with little to no warning. Despite the unpredictability of these events, your company can take measures to minimize or prevent the adverse effect that unexpected events can have on your network. Some companies, unfortunately, didn’t take precautions and suffered as a result. Let’s look at some of the worst network failures in recent history, ordered from most impacted users to the least, and how they could have been avoided.

1: Adobe Systems Data Breach

In this breach that occurred on October 3, 2013, the software giant claimed that around three million records had been stolen as a result of an orchestrated attack. However, the total number of user accounts accessed was around 38 million, compromising their personal information. Later, Adobe admitted to knowing about insufficient security measures surrounding their Cloud software offering. Boosting their security measures would have likely prevented this attack, and protected the information of millions of people.

2: SQL Slammer Worm

In January 2003, the SQL Slammer Worm ravaged the internet, overloading routers and bringing communications to an absolute crawl for most Internet users. The worm attacked through a security hole in Microsoft SQL Server and propagated itself like wildfire through the Internet, affecting over 90% of vulnerable hosts, at least 75,000 hosts, in 10 minutes. Financial services, airline systems, and government election systems, among others, suffered significant disruptions.

To make matters worse, the attack was completely avoidable - a security patch for the vulnerabilities had been available for six months. Unfortunately, many network admins were reluctant to install it because some of the patches that Microsoft had been pushing out around this time had caused server issues.

3: Code Red Worm

Similar to the SQL Slammer Worm, Code Red spread through another Microsoft security hole on July 16, 2001. The outbreak could have been prevented by using an available security patch. In this attack, the vulnerability was in Microsoft’s IIS Server. It affected web pages by defacing websites with erroneous text, attacking a host of websites, including Whitehouse.gov. Once again, routine server maintenance could have prevented this attack. This attack wasn’t quite as virulent as SQL Slammer, but impacted more important websites directly. Experts estimate that 359,000 hosts were affected in less than a half an hour.

4: Samsung Data Center Fire

A fire at a South Korean SDS subsidiary location in April 2014 disrupted Samsung products globally. While a natural disaster is not foreseeable, the scale of the after-effects could have been dramatically reduced. Unfortunately, Samsung had consolidated most of their mission-critical servers in one server room. This affected thousands of users of their electronic devices. Additionally, debit card and cash advance services of several financial institutions were totally brought offline by the fire. If their data center had been hosted in the cloud, or off-site and properly structured to failover to redundant locations, the outage could have been prevented or minimized.

5: GoDaddy.com DNS Corruption

On September 10, 2012, corrupted data in GoDaddy’s routing tables caused a widespread Internet outage. As one of the largest domain hosting companies, the GoDaddy outage caused millions of websites to be down for around six hours. The glitch allegedly resulted from aging hardware problems, which could have been prevented with cloud server hosting or virtualization.

6: Hurricane Sandy Natural Disaster in the Northeast

On October 29, 2012, Hurricane Sandy made landfall near Atlantic City, NJ. Flooding and widespread power outages affected data centers across the northeastern coast, causing over 2,000 networks to go offline. The storm led to many Internet providers rethinking their current failure strategies and resulted in a strengthening of infrastructure, increased migration to cloud hosting, and a redesign of emergency measures.

7: Hosting.com

On July 28, 2012, over a thousand websites hosted by Hosting.com went down because of botched maintenance on a UPS appliance - an employee shut down the power to their server racks causing outages for thousands of websites. This failure could have easily been prevented if Hosting.com had used a professional data center provider to manage their servers.

8: Internap Outage

The Internap data center in New York experienced power outage lasting long enough to drain the batteries of UPSs in May 2014. Internap provides live television broadcasts such as the Academy Awards and the Grammys, meaning that an outage of this magnitude affected a massive audience. Most of the company’s offerings were unavailable for almost seven hours. Had their failover plan been more robust, or if they had migrated their offerings to the cloud, the outage could have been avoided.

It’s clear that many of these issues could have been avoided with proper security measures, virtualizing servers, or shifting data to the cloud. If you want to keep your company’s data secure, no matter what unexpected events arise, get in touch with the team at Sanity Solutions. We’re happy to discuss the best ways to keep your network functioning and secure.


Comments