September 22, 2015

Would You Test Software Designed To Cheat On Tests?

Das Cheat?


The U.S. Environmental Protection Administration announced Friday that the Volkswagen AG had surreptitiously equipped its diesel vehicles with software designed to recognize when those products were being tested on a dynamometer, essentially an automotive treadmill. In such a situation, the full complement of emissions controls systems would operate at their maximum, bringing the vehicles into compliance with U.S. - and even tougher California - emissions standards.

But once the testing would be completed, according to the EPA, the vehicles would revert to a different mode, effectively allowing emissions levels to increase by as much as 40 times.

  • Volkswagen AG admitted that it built “defeat device” software into its diesel cars from 2009 to 2015 that automatically cheated on U.S. air-pollution tests
  • Software is built into 11 million vehicles worldwide
  • It could cost VW around $7.2 billion to fix the cars
  • Volkswagen's CEO resigned due to the scandal
  • VW designed the software to have the 3.0-liter engine run on two modes: “temperature conditioning” – the cheating mode – and “normal mode.”
  • The car reverted to normal mode “exactly one second” after the software detected that EPA testing conditions had stopped
  • VW has blamed a few rogue engineers for the scheme

It's pretty hard to believe "a few rogue engineers" would concoct and execute a scheme like this that, if discovered, could have catastrophic repercussions for the company. 

Imagine you worked for VW (or were a contractor), and were assigned to test this "defeat device" code. How would you feel? And how would you feel now that it's public knowledge?

See also:
http://www.foxnews.com/leisure/2015/09/22/volkswagen-says-11-million-its-cars-are-equipped-with-emissions-test-beating/

Update:

Apparently, VW's cheating was worse than originally reported, and extends to some Audi and Porsche models as well.

The EPA, California Air Resources Board and Environment Canada discovered the additional cheating while testing Volkswagen's vehicles. They have found nothing similar while conducting tests of other car companies.
Volkswagen engineers took a sneaky approach when fitting these vehicles with the "defeat device" software, the EPA alleged.

See Also:
http://www.usatoday.com/story/money/cars/2015/11/02/epa-diesel-suv-volkswagen-audi-porsche/75044132/
http://www.nytimes.com/2015/11/03/business/some-porsche-models-found-to-have-emissions-cheating-software.html?_r=0
http://www.latimes.com/business/la-fi-vw-action-20151102-story.html
http://www.consumeraffairs.com/news/epa-more-vw-diesels-have-defeat-device-software-110215.html

This article originally appeared in my blog: All Things Quality
My name is Joe Strazzere and I'm an experienced Quality Assurance professional.
I like to lead, to test, and occasionally to write about leading and testing.
Find me at http://AllThingsQuality.com/.

September 1, 2015

Perhaps They Should Have Tested More - SunGard


not the actual           BNY Mellon       logo *         

Perhaps SunGard and/or BNY Mellon should have tested more


If you've been watching the Market the past few weeks, you may have been concerned by all the volatility and heavy activity occurring.

And if your funds' NAV (Net Asset Value) was reported through BNY Mellon, you may have been even more concerned to learn that the NAVs at the end of the day's trading were delayed.

That's apparently because an operating system upgrade by SunGard (which hosts the fund account platform of BNY Mellon) caused a malfunction - on both the production platform, and the backup platform!

  • SunGard hosts the fund account platform of BNY Mellon
  • A malfunction in this platform disrupted the net asset values (NAVs) of BNY Mellon mutual funds and exchange-traded funds
  • Impacted 66 of BNY Mellon's US fund accounting clients
  • For up to 1200 funds valued at over $400 billion, the reporting of the NAVs at the end of the day's trading was delayed.
  • The timing was especially poor, due to the heavy market activity in the days that followed.
  • The problems may have led BNY Mellon to calculate inaccurate prices during the market's volatility
  • BNY Mellon's clients were likely receiving imprecise information
  • Per SunGard "issue appears to have been caused by an unforeseen complication resulting from an operating system change."
  • SunGard said it followed standard testing procedure for the implementation of this change, but the update still corrupted both BNY Mellon's primary and backup environments
  • BNY Mellon's chairman and CEO Gerald Hassell said the full restoration of SunGard's platform "has taken far longer than any of us would have expected."
  • Dozens of companies rely on Bank of New York data to determine the correct value for their funds
  • Bank officials say the problem began on Aug. 24, just as U.S. markets suffered their worst single-day loss since 2011

Perhaps it was just bad timing. 

It's nice that SunGard followed their standard testing procedures, but perhaps they weren't adequate. And perhaps it's not a good idea to make changes to the Backup environment, until you are completely sure the Production environment is actually working correctly.

Perhaps SunGard (and perhaps even BNY Mellon) should have tested more.

See also:


* actual BNY Mellon logo was removed per demand from BNY Mellon
"In connection to the Bank of New York Mellon Corporation’s proprietary rights over its famous trademarks we are notifying you of the following:

It has come to our attention that our trademark(s) appears as a metatag, keyword, visible or hidden text on the web site(s) located at:
http://www.allthingsquality.com/2015/09/perhaps-they-should-have-tested-more.html

without having obtained prior written authorization from The Bank of New York Mellon Corporation. This practice infringes upon our exclusive intellectual property rights."

This article originally appeared in my blog: All Things Quality
My name is Joe Strazzere and I'm an experienced Quality Assurance professional.
I like to lead, to test, and occasionally to write about leading and testing.
Find me at http://AllThingsQuality.com/.