RISKS-LIST: Risks-Forum Digest Wednesday 25 April 2012 Volume 26 : Issue 80 ACM FORUM ON RISKS TO THE PUBLIC IN COMPUTERS AND RELATED SYSTEMS (comp.risks) Peter G. Neumann, moderator, chmn ACM Committee on Computers and Public Policy ***** See last item for further information, disclaimers, caveats, etc. ***** This issue is archived at <http://www.risks.org> as <http://catless.ncl.ac.uk/Risks/26.80.html> The current issue can be found at <http://www.csl.sri.com/users/risko/risks.txt> Contents: Airline pilot distracted by new text messages botches landing attempt (Mike Flacy) Backdoor in mission-critical systems (Dan Goodin via C Y Cripps) The Power of Individual Voters to Transform Their Government (William Cox) Risks from computers in elections? (Brad Friedman via Mark E. Smith) Thieves steal fiber (Joel Garry) Berkeley High students hack into attendance system (Jill Tucker) TapLogger (Jim Reisert) Solar panel production page reveals name, address, real-time info (Jonathan Kamens) Re: Insider attack on smart meters (Paul Wallich) Compromised ATMs of no Consequence to Banks or Customers (Chris J Brady) Google Street View face blurring side effects (Ed Ravin) Occupy Wall Street protester doesn't own his tweets, judge rules (Lauren Weinstein) Harvard Library open access? (Dan Geer) White House pushes back on CISPA "cybersecurity" legislation (The Hill and LW) 11 percent of all sexts sent to the wrong recipient (Natt Garun via Monty Solomon) "The Flight from Conversation" (Sherry Turkle) Review: CERT Guide to Insider Threats, Capelli/Moore/Trzeciak (Ben Rothke) Henry Petroski does it again! (PGN) Gmail outage much broader than originally reported (Juan Carlos Perez via Gene Wirchenko) Re: "Did first DDOS attack sink the Titanic?" (Jeremy Ardley) Re: Hospital generator failure following earthquake (Dick Mills) The hidden danger of Windows 8 Microsoft Accounts (Woody Leonhard via Gene Wirchenko) Abridged info on RISKS (comp.risks) ---------------------------------------------------------------------- Date: Mon, 23 Apr 2012 17:14:02 -0400 From: Monty Solomon <monty_at_private> Subject: Airline pilot distracted by new text messages botches landing attempt (Mike Flacy) Mike Flacy, *Digital Trends*, 19 Apr 2012 While U.S. residents that fly commercially have to turn off their cell phones prior to take-off, a airline pilot in Australia left his phone on during a flight and found new text messages more interesting than landing the plane. As detailed by the Australian Transport Safety Bureau, an investigation into a Jetstar flight JQ57 between Darwin to Singapore discovered that the airline captain failed to lower the landing gear during the first attempt at a landing as he was too busy with his mobile phone. While the incident occurred nearly two years ago, the details of the investigation were released this week. According to the report, the captain neglected to turn off his mobile phone prior to the 220-seat Airbus 320 taking off in Darwin, Australia. When the plane began an initial descent into Changi Airport within Singapore, the captain's phone started beeping with new text message alerts when the plane was in between 2,500 to 2,000 feet off the ground. The captain turned his attention to the phone during the descent and the co-pilot attempted to get the captain's attention. After trying to alert the captain twice, the co-pilot switched off the auto-pilot during landing, but started to notice that something was wrong when the plane was just 1,000 feet off the ground. ... http://www.digitaltrends.com/mobile/airline-pilot-distracted-by-new-text-messages-botches-landing-attempt/ ------------------------------ Date: Wed, 25 Apr 2012 15:44:49 -0400 (EDT) From: C Y Cripps <cycmn_at_private> Subject: Backdoor in mission-critical systems (Dan Goodin) Backdoor in mission-critical hardware threatens power, traffic-control systems http://arstechnica.com/business/news/2012/04/backdoor-in-mission-critical-hardware-threatens-power-traffic-control-systems.ars Dan Goodin, *ArsTechnica", 25 Apr 2012 Backdoor in mission-critical hardware threatens power, traffic-control systems Like a key under a door matt, the MAC address exposed here allows hackers to tamper with this Internet-connected RuggedCom device, used to control power substations and other critical infrastructure. In the world of computer systems used to flip switches, open valves, and control other equipment inside giant electrical substations and railroad communications systems, you'd think the networking gear would be locked down tightly to prevent tampering by vandals. But for customers of Ontario, Canada-based RuggedCom, there's a good chance those Internet-connected devices have backdoors that make unauthorized access a point-and-click exercise. That's because equipment running RuggedCom's Rugged Operating System has an undocumented account that can't be modified and a password that's trivial to crack. What's more, researchers say, for years the company hasn't bothered to warn the power utilities, military facilities, and municipal traffic departments using the industrial-strength gear that the account can give attackers the means to sabotage operations that affect the safety of huge populations of people. "You treat these embedded appliances as a device that you don't have a window to see into," says researcher K. Reid Wightman of industrial machinery, which is often designed to withstand extreme heat and cold, dust, and other brutal conditions where they're housed. "You can't really patch it. You have to rely on the vendor to do the right thing when they set the device up and when they install the OS. And the vendor really fell down on this one." The backdoor uses the login ID of "factory" and a password that's recovered by plugging the MAC, or media access control, address of the targeted device into a simple Perl script, according to this post published on Monday to the Full Disclosure security list. To make unauthorized access easy, paying customers of the Shodan computer search engine can find the IP numbers of more than 60 networks that use the vulnerable equipment. The first thing users who telnet into them see, as the picture above demonstrates, is its MAC address. Like a router plugged into a utility's power grid Equipment running the Rugged Operating System act as the switches and hubs that connect programmable logic controllers to the computer networks used to send them commands. They may sit between the computer of a electric utility employee and the compact disk-sized controller that breaks a circuit when the employee clicks a button on her screen. To give the equipment added power, Rugged Operating System is fluent in the Modbus and DNP3 communications protocols used to natively administer industrial control and SCADA, or supervisory control and data acquisition, systems. The US Navy, the Wisconsin Department of Transportation, and Chevron are just three of the customers who rely on the gear, according to this page on RuggedCom's website. "As a citizen and based on the customer list on their website, I know for a fact that I personally depend on this equipment every day in some way," said Justin W. Clarke, the author of the full-disclosure advisory who said he notified company officials of the backdoor 12 months ago. "The equipment is so widely installed that it would be logical to assume that something I'm doing--whether it's riding a train, using power, or walking across a cross walk--depends on this." RuggedCom representatives didn't respond to a request for comment. This article will be updated if a response is received after its initial publication. According to a timeline included in Clarke's advisory, RuggedCom officials earlier this month stated "they need another three weeks to alert their customers, but not fix the vulnerability." Working with the US Computer Emergency Response Team, Clarke said he sought additional information, but RuggedCom never responded. Forever day bugs bite again In acknowledging but not fixing a security vulnerability in software that's widely used to control critical infrastructure, RuggedCom joins a growing roster of companies marketing wares bitten by so-called forever-day bugs. The term, which is a play on the phrase zero-day vulnerability, refer to documented flaws in industrial systems that will never be fixed. Other members of this group include ABB, Schneider Electric, and Siemens. Indeed, RuggedCom was acquired by a Canada-based subsidiary of Siemens in March. The hardcoded backdoor can be opened when users access affected devices using telnet, remote shell, or a serial console. The best defense against attacks that exploit the vulnerability is a layered approach that includes isolating devices from the Internet altogether as well as disabling or blocking telnet and remote shell access through network filters or firewalls, Clarke said. An independent security researcher in San Francisco, Clarke told Ars he has grown so concerned about the lack of security in industrial control systems that he's taken to ordering used gear hawked on eBay to see what kinds of vulnerabilities he can find in it. He said he spotted the Rugged OS backdoor with little trouble by analyzing an image of the RuggedCom firmware. "It is esoteric, it is obscure, but this equipment is everywhere," he said. "I was walking down the street and they had one of the traffic control cabinets that controls stop lights open and there was a RuggedCom switch, so while you and I may not see it, this is what's used in electric substations, in train control systems, in power plants and in the military. That's why I personally care about it so much." This article was updated to remove identifying information included in the image. [See also a WiReD item (noted by Lauren Weinstein): Rugged switches and servers are used in "mission-critical" communication networks used in power grids, railway control systems, and traffic control systems as well as in manufacturing facilities. RuggedCom asserts on its web site that its products are "the product of choice for high-reliability, high-availability, mission-critical communications networks deployed in harsh environments around the world." Clarke says he notified RuggedCom about his discovery in Apr. 2011 and says the representative he spoke with acknowledged the existence of the backdoor. "They knew it was there," he told Threat Level. "They stopped communicating with me after that." http://j.mp/Jqr8hm ] ------------------------------ Date: Tue, 24 Apr 2012 15:27:37 PDT From: "Peter G. Neumann" <neumann_at_private> Subject: The Power of Individual Voters to Transform Their Government William John Cox, Dandelion Salad: "U.S. voters appear to be increasingly powerless to fight the plutocracy which runs their government. As a result, Americans are living in an ever more repressive police state that is illegally committing acts of violent aggression around the world. The only thing that can possibly transform the U.S. government to one that cares for the voters who elect it, rather than for the plutocracy that controls it, is a unified opposition by all of the People, irrespective of their social class or political beliefs." http://truth-out.org/news/item/8705-the-power-of-individual-voters-to-transform-their-government ------------------------------ Date: Tue, 17 Apr 2012 19:58:10 -0700 From: "Mark E. Smith" <mymark_at_private> Subject: Risks from computers in elections? (Brad Friedman) A recent article by Brad Friedman: Baked Again in Alaska: Yet Another Election Crashes and Burns in The Last Frontier, states that, "...the electronic voting systems we use in this nation --- every single one of them --- are complete garbage." http://www.bradblog.com/?p=9259 One of the risks enumerated is that the central tabulators that count more than 90% of US votes (whether cast on electronic voting machines or on paper ballots read by optical scanners) can be remotely accessed without public awareness, and that data entered into them, particularly when all data comes from "admin" using the password "password," cannot be traced. Merely to learn that the data is not verifiable can take years of litigation, and the information sought can be "manipulated" before being released, if it is released at all. Ultimately, computer systems are only as reliable as the people who program and administer them. In the case of programmers working for voting machine corporations, and of politically appointed elections officials, the public usually has no way to assess their reliability. Most voters would not trust their lives and the lives of their children to the brakes on a car if they had no way to verify that the brakes were in good working condition. Yet they continue to trust their future and the future of their children and their country to an election system that is, indeed, "complete garbage," and, except in rare cases of hand counts, completely unverifiable. The real risk to the public isn't in the computer systems used in elections, which could easily be both accurate and verifiable, it is in voters placing their trust in systems that in more than 90% of cases cannot be verified as accurate, and in the people who program and administer those systems, whose reliability can rarely be assessed. Although Brad and many other voters continue to insist that nonvoters like myself are "apathetic," I think it is obvious that people who vote in systems they know to be unreliable, unverifiable garbage, are the ones who really don't care. ------------------------------ Date: Tue, 24 Apr 2012 17:31:33 -0700 From: joel garry <joel-garry_at_private> Subject: Thieves steal fiber Thieves stole 75 feet of fiber and 6 feet of copper cable. 600 strand cable served 10,000 Internet customers, including military bases, and some cellphone servers. http://www.utsandiego.com/news/2012/apr/24/thieves-steal-copper-and-fiber-optic-cables-alpine/ I'm not sure if the risk is that copper thieves are too stupid to know the difference between fiber and copper lines, or if criminal gangs now need fiber - they have been known to steal cellphone repeaters off of towers. An additional risk may be media magnification - TV reporting in the early morning said 17 million customers were affected. http://www.garry.to ------------------------------ Date: Fri, 20 Apr 2012 9:31:40 PDT From: "Peter G. Neumann" <neumann_at_private> Subject: Berkeley High students hack into attendance system (Jill Tucker) [Source: Jill Tucker, Berkeley High students hack into attendance system, *San Francisco Chronicle*, 20 Apr 2012; PGN-ed] http://www.sfgate.com/cgi-bin/article.cgi?f=3D/c/a/2012/04/20/MN0B1O5RPQ.DTL Some of Berkeley High's best students are among nearly three dozen students suspended for hacking into the school's attendance system. At least four students used an administrator's stolen password to clear tardies and unexcused absences from the permanent records of 50 students, offering the service or the password for a price. The scam allowed the students to circumvent the school's rigid attendance policy, which had been in effect until March 2012 and required teachers to dock student grades if they had three or more unexcused absences. The hackers erased from the system hundreds of cut classes and tardies from October through December 2011, and charged classmates $2 to $20 for the illicit assistance. ------------------------------ Date: Mon, 23 Apr 2012 15:59:33 -0600 From: Jim Reisert AD1C <jjreisert_at_private> Subject: TapLogger TapLogger: Inferring User Inputs On Smartphone Touchscreens Using On-board Motion Sensors http://www.cse.psu.edu/~szhu/papers/taplogger.pdf Similar to looking at fingerprint patterns on the screen to try to determine an unlock password. Jim Reisert AD1C, <jjreisert@private>, http://www.ad1c.us ------------------------------ Date: Wed, 18 Apr 2012 00:50:26 -0400 From: Jonathan Kamens <jik_at_private> Subject: Solar panel production page reveals name, address, real-time info The recent articles in RISKS about "smart meters" brought the following story to mind which might be of interest here... My wife and I had a solar photovoltaic electricity generation system, a.k.a. "solar panels," installed on our roof late last year by a Boston-area company called SunBug. Overall, we were very pleased with SunBug and with the quality of our installation, and we're looking forward to several decades of much, much lower electric bills (our last monthly bill was $1.27!). Many solar installers offer a web page their customers can use to monitor the performance of their systems. SunBug, however, goes one step further: their web page monitors not only solar electricity production, but also home electricity consumption. This allows customers to monitor how much of their electricity needs are being satisfied by solar, and it incentivizes them to find ways to conserve energy to make the line on the graph go down. It's a great idea in theory, but there are serious problems with SunBug's implementation of it. After our system went live, SunBug set up the monitoring page for us, sent us a link to it, and posted that link on a publicly accessible page on their web site. I.e., it was at that point possible for anyone in the world to access our monitoring page and view our production and consumption in real-time. But that's not all that was visible on the page; it also contained our name and address. I'm not going to post a link to our page because I don't know you all well enough to do that :-), but suffice it to say that any moderately intelligent person can tell from the consumption graph, with a high degree of accuracy, when we are home or away and awake or asleep. If someone wanted to rob us, they would be able easily to tell the best time to do that. Worse, if someone wanted to harm us or our children, they would be able easily to tell when we'd be available at home to be harmed. This is really bad. Their exposure of their customers' names and addresses on a publicly accessible web site without their prior knowledge or consent is clearly a violation of the Massachusetts Data Privacy Law, 201 CMR 17. There's another interesting twist in the story... As soon as I realized that our address was visible to the world on our monitoring page, I updated our settings to tell the site not to make our address visible. They do provide that as an option, although they don't enable it by default. However, after doing that, I was digging a little deeper into how the monitoring page works, and I discovered that the bit controlling whether to display the address is enforced on the client side via AJAX, not on the server. In other words, our address was being sent down to the web browser regardless of whether the bit was set; all the setting did was tell the browser whether to display the address to the user. Therefore, anyone could use a network sniffer like wireshark or even just a browser tool like Firebug to find out our address. Needless to say, I raised a big stink to SunBug. They fixed it for us by making back-end changes in their database to completely remove all of our PII from the site. They also said they understood my concerns and were looking into how to address them for all of their customers, although I don't know what in particular they've done in the several months since I brought the issue to their attention. The box that SunBug installed to monitor our electricity production and consumption is a "smart meter" of sorts, so this story illustrates that the concerns people have with smart meters are legitimate and already manifesting in the real world. In closing, I want to emphasize that although I was rather disappointed with this particular aspect of SunBug's work, I was pleased overall, and I still think we chose the correct installer out of the five or so we interviewed seriously before hiring SunBug. [And of course the OpSec solution is decidedly anti-ecological: make consumption appear exactly the same all the time! ------------------------------ Date: Tue, 17 Apr 2012 21:07:59 -0400 From: Paul Wallich <pw_at_private> Subject: Re: Insider attack on smart meters (RISKS-26.79) > FBI Concerned About Smart Meter Hacking, 9 Apr 2012 > http://krebsonsecurity.com/2012/04/fbi-smart-meter-hacks-likely-to-spread/ I am a little suspicious of the numbers here (and, in the original report they appear to be hedged with subjunctives). A few searches show that Puerto Rico consumes about 22 billion kWh annually, with a typical retail cost of roughly 10 cents/kWh. So to get into the hundreds of millions (the linked piece cites an estimate of "up to" $400M/yr) you would need to steal 10-20% of total electricity consumption on the island. Even assuming that basic software checks would ignore a drop of as much as 75% in consumption year-over-year, that would require about 15-25% (and possibly more) of total consumption to be through hacked meters. You would need a large industrial enterprise to do that kind of work. Anyone want to bet that the "hundreds of millions" is an estimate of what might be the losses if all the smart meters were hacked? ------------------------------ Date: Mon, 23 Apr 2012 03:23:06 -0700 (PDT) From: Chris J Brady <chrisjbrady_at_private> Subject: Compromised ATMs of no Consequence to Banks or Customers In London many ATMs have been targeted by criminal gangs for years. And it appears that no-one cares -- many ATMs are visibly compromised -- yet the banks allow them to remain damaged. I know of three; 1/ one in Charing Cross Station with wires hanging out of the fascia, 2/ one in Paddington Station where the whole machine can be pulled out of the cabinet, 3/ one at the top end of Charing Cross Road with glue all round from where a skimming device had been installed. But the situation is worse. At Paddington I was using another -- undamaged -- ATM. Afterwards I saw someone using the damaged ATM (she had the obligatory earbuds in both ears), so after she had finished I pointed out that the ATM was damaged and its security compromised. She didn't care. She walked off. Ditto with some users of the ATM at Charing Cross. No-one could be bothered to listen or take notice. Late last year I reported the one at Charing Cross to Barclays Bank - it remains compromised 5 months later. And this will get worse as the tourists arrive for the Olympics. ------------------------------ Date: Sun, 22 Apr 2012 18:44:54 -0400 From: Ed Ravin <eravin_at_private> Subject: Google Street View face blurring side effects The Israeli daily HaAretz reports that Google Street View is now live for several major Israeli cities. The Justice Ministry demanded that Google blur out faces and license plates, but there's still a few bugs in the system - faces on some posters and advertisements were blurred, but some faces on actual people were not: http://www.haaretz.com/business/google-street-view-goes-live-in-israel-2-days-early-1.425753 ------------------------------ Date: Tue, 24 Apr 2012 22:02:30 -0700 From: Lauren Weinstein <lauren_at_private> Subject: Occupy Wall Street protester doesn't own his tweets, judge rules In a candid ruling, a New York judge said a protester can't stop prosecutors from searching his Twitter account because he doesn't own the tweets in the first place. Judge Matthew Sciarrino Jr. cited a "widely-believed" but "mistaken" notion about online privacy rights and said that search and seizure protections don't apply because we "do not have a 'physical' home on the Internet." http://j.mp/ID36Rk (Paid Content) [T'weet or not T'weet, That is di-gestion. PGN] ------------------------------ Date: Tue, 24 Apr 2012 10:03:59 -0400 From: Dan Geer <dan_at_private> Subject: Harvard Library open access? http://boingboing.net/2012/04/23/harvard-library-to-faculty-we.html Harvard Library to faculty: we're going broke unless you go open access Cory Doctorow, Monday, Apr 23 "Harvard Library's Faculty Advisory Council is telling faculty that it's financially 'untenable' for the university to keep on paying extortionate access fees for academic journals. It's suggesting that faculty make their research publicly available, switch to publishing in open access journals and consider resigning from the boards of journals that don't allow open access." ------------------------------ Date: Tue, 17 Apr 2012 18:42:01 -0700 From: Lauren Weinstein <lauren_at_private> Subject: White House pushes back on CISPA "cybersecurity" legislation "In a statement, National Security Council spokeswoman Caitlin Hayden said any cybersecurity legislation should include strong privacy protections and should set mandatory security standards for critical infrastructure systems, such as electrical grids and water supplies." http://j.mp/IWlFmS (The Hill via NNSquad) My current blog posting re CISPA is: " "CISPA, Cybersecurity, and the Devil in the Dark": http://j.mp/HIO8ud (Lauren's Blog) Lauren Weinstein (lauren@private): http://www.vortex.com/lauren http://lists.nnsquad.org/mailman/listinfo/nnsquad ------------------------------ Date: Mon, 23 Apr 2012 17:18:28 -0400 From: Monty Solomon <monty_at_private> Subject: 11 percent of all sexts sent to the wrong recipient (Natt Garun) Natt Garun, *Digital Trends*, 19 Apr 2012 Sexters beware: A new poll finds that your sexy messages may end up in the wrong hands if you're not careful enough. If you're into mobile, virtual sexy time, we have no judgment with what you want to do in your private life. But according to a poll conducted by United Kingdom-based mobile news site Recombu, 11 percent of sexts are sent to unintended recipients. Looks like too many of you are getting too caught up in the moment! The poll, which surveyed approximately 2,000 adults, showed that 47 percent of responders sext on a regular basis. About 48 percent of sexters are female, and 45 percent are male (we're guessing the rest means undisclosed gender or transgendered). However, the numbers show that males seem to get more heated and eager than their female counterparts. About one in 10 sexts from male senders get accidentally shipped to someone it wasn't meant for, while females stats stand at one in 20. What's worse: 16 percent of men have had their sexy messages end up in a family member's inbox while just 8 percent of females suffer the same embarrassment. ... http://www.digitaltrends.com/mobile/11-percent-of-all-sexts-end-up-sent-to-the-wrong-recipient/ ------------------------------ Date: Mon, 23 Apr 2012 9:49:23 PDT From: "Peter G. Neumann" <neumann_at_private> Subject: "The Flight from Conversation" (Sherry Turkle in The NYT) [Source: Sherry Turkle, The Flight From Conversation, *The New York Times*, The Sunday Review, 22 Apr 2012; excellent long article, PGN-ed] http://www.nytimes.com/2012/04/22/opinion/sunday/the-flight-from-conversation.html We live in a technological universe in which we are always communicating. And yet we have sacrificed conversation for mere connection. At home, families sit together, texting and reading e-mail. At work executives text during board meetings. We text (and shop and go on Facebook) during classes and when we're on dates. My students tell me about an important new skill: it involves maintaining eye contact with someone while you text someone else; it's hard, but it can be done. ... I spend the summers at a cottage on Cape Cod, and for decades I walked the same dunes that Thoreau once walked. Not too long ago, people walked with their heads up, looking at the water, the sky, the sand and at one another, talking. Now they often walk with their heads down, typing. Even when they are with friends, partners, children, everyone is on their own devices. So I say, look up, look at one another, and let's start the conversation. Sherry Turkle is a psychologist and professor at M.I.T. and the author, most recently, of ``Alone Together: Why We Expect More From Technology and Less >From Each Other.'' [This really gives new meaning to being left to your own devices. PGN] ------------------------------ Date: Wed, 18 Apr 2012 23:20:27 -0400 From: Ben Rothke <brothke_at_private> Subject: Book review: CERT Guide to Insider Threats Dawn Cappelli, Andrew Moore and Randall Trzeciak CERT Guide to Insider Threats: How to Prevent, Detect, and Respond to Information Technology Crimes CERT Insider Threat Center While Julius Caesar likely never said ``Et tu, Brute?'', the saying associated with his final minutes has come to symbolize the ultimate insider betrayal. In The CERT Guide to Insider Threats: How to Prevent, Detect, and Respond to Information Technology Crimes, authors Dawn Cappelli, Andrew Moore and Randall Trzeciak of the CERT Insider Threat Center provide incontrovertible data and an abundance of empirical evidence, which creates an important resource on the topic of insider threats. There are thousands of companies that have uttered modern day versions of Et tu, Brute due to insidious insider attacks and the book documents many of them. The book is based on work done at the CERT Insider Threat Center, which has been researching this topic for the last decade. The data the threat center has access to is unparalleled, which in turn makes this the definitive book on the topic. The threat center has investigated nearly 1,000 incidents and their data sets on the topic are unrivaled. With that, the book truly needs to be on the desktop of everyone tasked with data security and intellectual property protection. Full review at: http://365.rsaconference.com/blogs/securityreading/2012/04/18/the-cert-guide-to-insider-threats-how-to-prevent-detect-and-respond-to-information-technology-crimes ------------------------------ Date: Wed, 25 Apr 2012 15:22:07 PDT From: "Peter G. Neumann" <neumann_at_private> Subject: Henry Petroski does it again! Henry Petroski is a long-time friend of RISKS, and his name has appeared in at least two dozen RISKS items over the years. As I write this, I have just noticed that his latest regular column (this one is called Backseat Designers) in the May-June issue of *American Scientist* is adapted from Chapter 11 of his latest book, To Forgive Design: Understanding Failure, Belknap Press of Harvard University, 2012. Many years ago, when I managed to induce Henry to give keynote talks for an early COMPASS conference on safety and for the ACM SIGSOFT conference in 1991, he modestly insisted he did not know enough about computers. However, much of what he has written and said is always highly relevant to RISKS, and therefore his new book deserves mention here. ------------------------------ Date: Thu, 19 Apr 2012 19:35:22 -0700 From: Gene Wirchenko <genew_at_private> Subject: Gmail outage much broader than originally reported (Juan Carlos Perez) Juan Carlos Perez, *InfoWorld*, 18 Apr 2012 Google initially underestimated the number of affected users, which likely topped 30 million https://www.infoworld.com/d/applications/gmail-outage-much-broader-originally-reported-191180 ------------------------------ Date: Thu, 19 Apr 2012 14:21:26 +0800 From: Jeremy Ardley <jeremy.ardley_at_private> Subject: Re: "Did first DDOS attack sink the Titanic?" This may perhaps have been the first DDOS, but remember the telephone was around before then. Today we have the massive problem of emergency services numbers being overwhelmed by calls about a single incident. A lot of this is due to mobile phones but some exists because of the size of the incident. A typical scenario is a grass-fire besides a freeway. In the space of a few minutes hundreds of people will call in to report it. This massive volume of calls effectively prevents any response to other calls like a house fire. In the event of 9/11 People all over the city reported in. I literally have no idea how the call-centres coped with this massive amount of incoming calls and effectively a fixed number of operators. Back in 1900 I don't know if they had emergency numbers or just operators, but I'm sure any significant event would have overwhelmed whatever systems they had in place. The industry has yet to design a resilient call response system that can handle peak overloads while still attending to routine but life critical calls. ------------------------------ Date: Wed, 18 Apr 2012 13:52:38 -0400 From: Dick Mills <dickandlibbymills_at_private> Subject: Re: Hospital generator failure following earthquake RISKS-26.79 described how an earthquake stirred up sediments in diesel fuel tanks, causing the emergency diesel engines to stop. This problem is well known to boaters. The classical scenario is trying to escape from an enclosed harbor to the open sea. The pounding of the boat in the surf at the harbor entrance stirs sediments, predictably causing engine failure at the worst possible time. The known remedy is to have multiple fuel filters plus a rapid way to switch from one to the backup. Knowledgeable skippers in that classic scenario will even station a man to stand by with his hand on the fuel filter throw-over valve. Even in less classic scenarios, it is completely foreseeable that fuel filters are most likely to clog at the time of maximum agitation. Since that is precisely the time when dependence on engine power is most critical, it must be addressed at the design and planning stages. In an unmanned installation, the switching filters remedy is problematic (even if automatic). Still, the nature of the problem is identical. So the risk here involves mechanisms to transfer risk knowledge from one domain to another. Surely engine manufacturers must be familiar with it. I wonder what their installation and operating instructions say on the subject. Dick Mills, Sailing Vessel Tarwathie ------------------------------ Date: Wed, 18 Apr 2012 14:22:43 -0700 From: Gene Wirchenko <genew_at_private> Subject: The hidden danger of Windows 8 Microsoft Accounts (Woody Leonhard) Woody Leonhard, *InfoWorld*, 18 Apr 2012 Microsoft goes to great lengths to convince Windows 8 users to log on with an email address, but if your account gets hijacked you could find yourself locked out http://www.infoworld.com/t/microsoft-windows/the-hidden-danger-of-windows-8-microsoft-accounts-191185 ------------------------------ Date: Mon, 6 Jun 2011 20:01:16 -0900 From: RISKS-request_at_private Subject: Abridged info on RISKS (comp.risks) The ACM RISKS Forum is a MODERATED digest. Its Usenet manifestation is comp.risks, the feed for which is donated by panix.com as of June 2011. => SUBSCRIPTIONS: PLEASE read RISKS as a newsgroup (comp.risks or equivalent) if possible and convenient for you. The mailman Web interface can be used directly to subscribe and unsubscribe: http://lists.csl.sri.com/mailman/listinfo/risks Alternatively, to subscribe or unsubscribe via e-mail to mailman your FROM: address, send a message to risks-request_at_private containing only the one-word text subscribe or unsubscribe. You may also specify a different receiving address: subscribe address= ... . You may short-circuit that process by sending directly to either risks-subscribe_at_private or risks-unsubscribe_at_private depending on which action is to be taken. Subscription and unsubscription requests require that you reply to a confirmation message sent to the subscribing mail address. Instructions are included in the confirmation message. Each issue of RISKS that you receive contains information on how to post, unsubscribe, etc. => The complete INFO file (submissions, default disclaimers, archive sites, copyright policy, etc.) is online. <http://www.CSL.sri.com/risksinfo.html> The full info file may appear now and then in RISKS issues. *** Contributors are assumed to have read the full info file for guidelines. => .UK users may contact <Lindsay.Marshall_at_private>. => SPAM challenge-responses will not be honored. Instead, use an alternative address from which you NEVER send mail! => SUBMISSIONS: to risks_at_private with meaningful SUBJECT: line. *** NOTE: Including the string "notsp" at the beginning or end of the subject *** line will be very helpful in separating real contributions from spam. *** This attention-string may change, so watch this space now and then. => ARCHIVES: ftp://ftp.sri.com/risks for current volume or ftp://ftp.sri.com/VL/risks for previous VoLume http://www.risks.org takes you to Lindsay Marshall's searchable archive at newcastle: http://catless.ncl.ac.uk/Risks/VL.IS.html gets you VoLume, ISsue. Lindsay has also added to the Newcastle catless site a palmtop version of the most recent RISKS issue and a WAP version that works for many but not all telephones: http://catless.ncl.ac.uk/w/r <http://the.wiretapped.net/security/info/textfiles/risks-digest/> . ==> PGN's comprehensive historical Illustrative Risks summary of one liners: <http://www.csl.sri.com/illustrative.html> for browsing, <http://www.csl.sri.com/illustrative.pdf> or .ps for printing is no longer maintained up-to-date except for recent election problems. ==> Special Offer to Join ACM for readers of the ACM RISKS Forum: <http://www.acm.org/joinacm1> ------------------------------ End of RISKS-FORUM Digest 26.80 ************************Received on Wed Apr 25 2012 - 16:25:06 PDT
This archive was generated by hypermail 2.2.0 : Wed Apr 25 2012 - 16:58:49 PDT