The December 2022 issue of IEEE Spectrum is here!

Close bar

San Francisco's BART System Went Down Due to Server Upgrade Gone Bad

Amtrak train gets lost, Boeing tells operators of 787 and 747-8 aircraft to avoid thunderstorms

3 min read
San Francisco's BART System Went Down Due to Server Upgrade Gone Bad
Maurits90/Wikipedia

IT Hiccups of the Week

Once more with feeling: the mêlée involving the Affordable Care Act website yet again dwarfed last week’s other IT-related impediments, which were relatively few for a change.

During last week’s round-the-clock Obamacare website glitch watch, for instance, we heard a government official admit that somewhere around 30 percent to 40 percent (no one seems to know for certain) of the required ACA back-office computing functionality related to how insurance companies get paid hasn’t been built yet. Documents were revealed showing that senior Obama Administration officials were worried, just before the website’s roll out, that there could be major problems—even though these same officials have claimed they had no inkling that website’s operation would lay down and play dead once going live. It was also revealed that, in a load test conducted just days before the website went live, the system choked when 500 users attempted to access the website simultaneously.  We also heard the Administration redefine operational success: a website that would work smoothly for 80 percent who try to enroll. This was immediately followed by debates about what that 80 percent measure actually means—if anything other than that a lot of people won’t be able to enroll for ACA health insurance via the website despite the promise of an “optimally functioning” website that would “work smoothly” by the end of November. These events had more than a little bit to do with extensions to the ACA 2014 and 2015 enrollment periods in order to help meet both Administration technical and political objectives. HealthCare.gov had company in its misery: There were continued delays to CuidadoDeSalud.gov, the Spanish-language version of the ACA website. Finally, despite everything, the White House released an upbeat report assuring the nation that everything will indeed soon be fine.  

The other IT-related obstacles, impairments and nervous breakdowns of the week included two rail system uffdas—one computer-related, and one apparently mechanical-cum-human error related. The first concerns a service outage on San Francisco’s BART (Bay Area Rapid Transit) system that lasted from late Thursday night into Friday morning. It was apparently caused by a server upgrade Thursday night that didn’t go according to plan. The second rail outage involved a New York City-bound Amtrak train that ended up going to Bala Cynwyd, Pennsylvania (outside Philadelphia) instead.

Finally, Boeing warned the 15 operators of Boeing’s 787 Dreamliner and 747-8 jumbo aircraft equipped with GEnx engines by GE not to fly at high attitude within 50 nautical miles of thunderstorms that may contain ice crystals. Apparently, there’s a risk of engine icing problems. Boeing and GE say that they are looking at a software fix to the engine control system which should be available early next year.

San Francisco’s BART System Goes Down for Several Hours

BART System Restored, But Commuters Left Seething

Software Problems Blamed for BART System Outage Trapping a Thousand Passengers

BART Explains Outage Caused by Bad Upgrade to Network Server

Amtrak Train 664 to New York City Ends Up in Philadelphia Suburb

Amtrak Train Crew Misreads Signal, Gets Lost

Amtrak Gets Turned Around on Way to New York City

Train Mechanical Problem Leads to Human-Error on Lost Amtrak Train

Boeing Tells 787 Dreamliner and 747-8 Jumbo Operators to Avoid Thunderstorms

Six Boeing Aircraft With GEnx Engines Have Had Engine Icing Problems

Boeing Issues Ice Risk Warning for GE-Powered 787 and 747-8 Aircraft

JAL Pulls 787 Off Two Routes

Of Other Interest …

Glitch delays 7500 Hennepin County Minnesota Employee Paychecks

Property Taxes Doubled In Princeton New Jersey Due to Software Glitch

New Election System Fails in Swaziland

Barclays Bank UK Online Systems Goes Out

Technical Glitch Takes Down Mexico Stock Exchange

Technical Glitch Blamed for Trading Halt on Qatar Exchange

Tesco Pricing Glitch Allows £9 Wine to Sell for £2.75

Restaurant Reputations in Northern Colorado Tainted by Health Department Software Error

Emergency Response System at SF Airport Failed Due To Software Problem Soon After July Crash

 

Photo: Maurits90/Wikipedia

The Conversation (0)

Why Functional Programming Should Be the Future of Software Development

It’s hard to learn, but your code will produce fewer nasty surprises

11 min read
Vertical
A plate of spaghetti made from code
Shira Inbar
DarkBlue1

You’d expectthe longest and most costly phase in the lifecycle of a software product to be the initial development of the system, when all those great features are first imagined and then created. In fact, the hardest part comes later, during the maintenance phase. That’s when programmers pay the price for the shortcuts they took during development.

So why did they take shortcuts? Maybe they didn’t realize that they were cutting any corners. Only when their code was deployed and exercised by a lot of users did its hidden flaws come to light. And maybe the developers were rushed. Time-to-market pressures would almost guarantee that their software will contain more bugs than it would otherwise.

Keep Reading ↓Show less
{"imageShortcodeIds":["31996907"]}