Brisbane airport shut down due to radar failure
Air traffic at Brisbane Airport was disrupted for a duration of about 40 minutes earlier today:
A data processing unit fault at Brisbane Airport had triggered an alert, with all flights in and out of Brisbane halted as a result.
Incoming flights from places such as Fiji and Coffs Harbour were diverted to Coolangatta Airport or put in a holding pattern over Brisbane for almost 40 minutes.
The problem has since been fixed and regular traffic control restored.
News of Michael Jackson’s death slows down major web sites
As news of Michael Jackson’s death broke yesterday afternoon, numerous major websites and services buckled under the onslaught of traffic:
Google was not the only company overwhelmed by the public’s clamour for information.
The microblogging service Twitter crashed with the sheer volume of people using the service.
[...]
TMZ, the popular celebrity gossip site that broke the story following a tip-off that a paramedic had visited the singers home also crashed.There was a domino effect as users then fled to other sites. Hollywood gossip writer Perez Hilton’s site was among those to flame out.
Keynote Systems reported that its monitoring showed performance problems for the web sites of AOL, CBS, CNN, MSNBC and Yahoo.
Wikipedia also experienced problems, as reported by CNET:
By around 3:15 p.m. PDT, Wikipedia appeared to be temporarily overloaded. The site reported the error: “Sorry! This site is experiencing technical difficulties… Cannot contact the database server: Unknown error (10.0.6.24))”
This raises a very valid question: Can the Internet handle big breaking news?
The statistics are amazing: Akamai said worldwide Internet traffic was 11 percent higher than normal during the peak hours between 3 p.m. PDT and 4 p.m., when news of Jackson’s death was breaking. That traffic forced even Google to its knees for a brief period of time Thursday afternoon.
The debate is an important one and it will likely be conducted at numerous places in the future.
Notably, the traffic increase was just 11 percent. As one source of information became unavailable, people moved on to other sources, which created a traffic ripple effect. It makes me wonder, what behavior could be observed, if several large news events occurred at roughly the same time.
comp.risks and related
Peter G. Neumann has been moderating the comp.risks usenet group. The digest is available here. Some of that and related content has also made it into the Sigsoft’s Software Engineering Notes‘ Risks to the Public column as well as ACM’s Inside Risks.
Excellent resources on numerous relevant cases.
Cashless payment system breakdown at The Great American Food and Music Fest
The Great American Food and Music Fest at Shoreline Amphitheatre in Mountain View, CA turned out to be a fiasco for the organizers and many of the visitors earlier this month. As Mercury News reports:
Saturday’s concert at the Shoreline Amphitheatre — the second problem-plagued event at the Mountain View venue in less than a month — also left organizers with more than heartburn. They are scrambling to deal with thousands of complaints from customers who had to stand up to five hours in line for food and endure a meltdown of a high-tech electronic bracelet pay system.
Numerous factors contributed to this becoming an overall bad experience for a lot of visitors. Larger-than-expected numbers of people started coming in early in the day. Food vendor lines were often very long and poorly organized. Vendors did not have enough food. The electronic payment system failed.
The organizers are now struggling with the aftermath, issuing apologies and refunds. Apparently the event will receive a second chance next year.
Winnipeg Parking Authority wrongfully issued parking notices to some 800
Some 800 people in Winnipeg, Canada were wrongfully sent parking notices by Winnipeg Parking Authority in the spring. As reported by cbc.ca:
A computer glitch resulted in hundreds of Winnipeggers receiving notices of unpaid parking violations — even though they had not been ticketed.
The notices were issued in the spring by the Winnipeg Parking Authority and sent to some 800 people. Colin Stewart, spokesman for the parking authority, said the problem was corrected and nobody was charged.
But nobody was notified, either.
There were really at least two computer-related issues here. This was not a result of software testing like in the Chicago case. Rather it sounds like incorrect information was applied to a numerous customer accounts. Then the staff at WPA had apparently no easy way to undo the change, nor figure out which records were affected. The latter could have allowed them to proactively inform people.
Electing to simply wait for incoming complaints really just puts the burden for WPA’s error on those, whose records were affected by this.
Some Pennsylvania Lottery players unable to cash in their winnings
A small percentage of people were unable to cash in their winnings from Pennsylvania Lottery this week, as reported by Times Leader:
Veronica Anderson, spokesperson for the Pennsylvania Lottery, said there was a “minor glitch” in the computer system for the June 17th Powerball drawing.
Anderson said the glitch occurred when a software conversion took place that day.
Approximately 200 out of 87,000 winning Powerball tickets were affected, Anderson said.
“Retailers were aware of the conversion,” Anderson said. “We experienced a minor problem affecting 200 tickets that weren’t in the system.”
Playing the Lottery is a losing proposition for the majority of players. It must be particularly frustrating to be prevented from cashing in, if you do actually win something – no matter how little.
Software testing results in over 100 parking tickets
Tom Feddor, whose Illinois license plate is “0″ has received more than 100 ticket notices from the City of Chicago, as reported by the Chicago Tribune:
A glitch occurred at the Chicago Department of Revenue involving Feddor’s 0 plates being used during tests of ticketing equipment. The error prompted the cascade of ticket notices to land in his mailbox, city officials determined after launching an internal investigation based on calls from the Tribune.
Apparently, the City of Chicago used Feddor’s license plate number to perform tests on their electronic ticket issuing system:
It turned out that some city parking-enforcement aides punched in 0 when testing their electronic ticket-issuing devices, Revenue Department spokesman Ed Walsh said. Officials weren’t aware there was a 0 plate or that Feddor was receiving tickets, Walsh said in response to the Tribune inquiry.
“The test violations should have been dismissed in the database. The majority of the cases [Feddor] contested successfully. But we are taking steps to rectify the situation so in the future an actual registered plate number will not be used to do the testing,” Walsh said.
There seem to be a number of risky technical issues here, such as as performing testing on live customer data, not checking whether the plate is currently in use, and of course actually issuing tickets to the customer.
Found via Near Loop Wire.
Fumes released at Holiday World theme park
The Chicago Tribune reports a filter malfunction at Holiday World, a holiday-inspired water park in Indiana this past Saturday. Numerous people suffered from breathing problems due to the fumes:
A filter pump malfunction in a water ride at southern Indiana’s Holiday World caused fumes that sent 42 people to a hospital with breathing troubles.
Park president Will Koch (COOK) says the pump sent highly concentrated levels of bleach and muriatic acid into the Bahari River waters at Splashin’ Safari on Saturday.
Life guards immediately got people out of the water and closed off the area. Some customers required first aid, others drove to the hospital or were taken there by ambulance. The issue has since been fixed and the ride re-opened.
MIT Technology Review’s 2002 article on 10 Technology Disasters
MIT Technology Review published an article 10 Technology Disasters back in 2002:
What do a 17th-century Swedish warship, an opulent Chicago theater and a Kansas City hotel “skyway” have in common? All met catastrophic ends–and they have important lessons to teach today’s innovators.
Here is a list of the examined examples, along with reference links to Wikipedia (except for the AT&T crash).
- The Vasa sinking
- The Hyatt Regency walkway collapse
- The Iroquois Theater blaze
- The Eschede train derailment
- The Ashtabula Creek Bridge wreck
- The St. Francis Dam burst
- The Atlantic Empress/Aegean Captain collision
- The AT&T network crash
- The 1965 Northeast blackout
- The Concorde crash
We’ll take more in-depth looks at some of these and many other examples in future posts and articles.
Phone billing errors in South Devon
Post Office Telecom Services sent out numerous erroneous phone bills to households in South Devon, England, according to this report:
Mr Postlethwaite of Lee Court, Centry Road, Brixham, had been a BT customer until recently switching to Talk Talk.
He said: “I have always paid my bills and, in fact, when I left BT I had a cheque back from them because I was in credit. I had always paid my bills on time.”
Mr Postlethwaite says he has now been told it was a computer error and thousands of similar incorrect bills have gone out.
From the article it is hard to tell, whether non-customers received bills, too. No action needs to be taken and the bills can be discarded. It looks like the fault (and expense) lies entirely with the phone company and their computer system.
