precedence: bulk Subject: Risks Digest 21.46 RISKS-LIST: Risks-Forum Digest Tuesday 12 June 2001 Volume 21 : Issue 46 FORUM ON RISKS TO THE PUBLIC IN COMPUTERS AND RELATED SYSTEMS (comp.risks) ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator ***** See last item for further information, disclaimers, caveats, etc. ***** This issue is archived at and by anonymous ftp at ftp.sri.com, cd risks . Contents: Another NY Stock Exchange outage (PGN) California power grid hacked (PGN) PC parrot drives firemen crazy (Merlyn Kline) Computer reports unreported wreck (Chris Norloff) U.K. plans mandatory IP indoctrination for children (Cluebot via Declan McCullagh) Re: Billboard error message (Robert Meineke, Rick Prelinger, John Dallman) Re: Risks of clueless marketing (Jamie McCarthy) Re: Steve Gibson: Windows XP Vulnerable; Big ISPs just don't care (Mike Nuss) Re: Steve Gibson's report and Windows XP "Vulnerabilities" (David Crooke) They're at it again: Internet Explorer Smart Tags in WinXP (Stef Maruch) Re: Office XP modifies what you type (Andy Newman, Jay Jennings) Microsoft, 'Mitigating Factors' and Public Relations (Jackson Ratcliffe) Broken shopping carts (Steve Loughran) How to avoid Internet interruption at AAS meeting (Clive Page) There's no such thing as software `piracy' (Fred Gilham) Re: Another fear of Risks (James K. Huggins) Re: McDonald's testing cashless payments (Jeffrey Jonas, John R Levine) Credit where it isn't due (William Paul Fiefer) Abridged info on RISKS (comp.risks) ---------------------------------------------------------------------- Date: Fri, 8 Jun 2001 19:21:22 PDT From: "Peter G. Neumann" Subject: Another NY Stock Exchange outage A software upgrade glitch resulted in the New York Stock Exchange being unable to trade roughly half of its stocks in the morning of 8 Jun 2001. Consequently, the exchange was shut down entirely (on grounds of fairness) until 11:35 a.m. EDT. The RISKS archives note a 41-minute shutdown on 24 Feb 1971 (when both primary and backup systems failed), a 24-minute outage on 22 Oct 1991 (due to a power dip), a one-hour outage on 18 Dec 1995 (also due to a botched software update), and a one-hour crash on 26 Oct 1998. Uninterrupted service is clearly not easy to achieve. The Nasdaq exchange computer system also shut down last week for 20 minutes (while the staff was working to increase capacity), a case that has not previously been reported here. ------------------------------ Date: Tue, 12 Jun 2001 08:13:22 -0700 From: "Peter G. Neumann" Subject: California power grid hacked Reuters reported on 11 June 2001 that the California Independent System Operator's flow-control computer systems had been hacked for at least 17 days before it was detected on 11 May 2001 -- in the midst of the ongoing power crisis. Although they attacks did not noticeably disrupt operations, they apparently came quite close -- and exposed some vulnerabilities that demonstrably need to be fixed. The main attack was seemingly from someone in China's Guangdong province, via China Telecom, and exploited Internet servers in Tulsa OK and Santa Clara CA. ------------------------------ Date: Thu, 7 Jun 2001 13:08:17 +0100 From: "Merlyn Kline" Subject: PC parrot drives firemen crazy In an article in *The Register*, Kieren McCarthy reported that West Midlands firemen, having rescued a cat from a tree, were called to an office in Willenhall to rescue what was thought to be an escaped parrot. After an hour's search, they discovered that a PC screensaver was intermittently parroting a parrot's squawks. Kieren speculated on whether the firemen thought it was a joke or "more reasonably, smashed the PC to pieces with their axes." [Merlyn called this a "terrible parroty error", although I doubt that the firemen thought it was a parody. Instead, it was truly a case of a polly-morphic PC! PGN-ed] ------------------------------ Date: Thu, 7 Jun 2001 08:44:52 -0400 From: "Chris Norloff" Subject: Computer reports unreported wreck You just can't outrun a satellite. A Merced, California, man took his fully equipped 2001 SUV out onto some nearby country roads, navigating swiftly and confidently with the optional OnStar Global Positioning System. When he got into an accident, he decided to run for it. But the guidance system had already notified OnStar headquarters of the accident, specifying where it had happened and giving a complete description of his vehicle to the California Highway Patrol. The officers followed a trail of coolant about a mile into an orchard, where they found and arrested the driver. [Source: *Road & Track* magazine, July 2001; PGN-ed] THE RISKS? What constitutes an "accident"? (Air bags seem to go off quite easily, taking out the windshield and dashboard [$$$] in a fender-bender). Will GPS-reported accidents become like household burglar alarms - sending out mostly false alarms? Who will hack into the OnStar system to falsely report accidents? Who will use the OnStar system to efficiently dispatch lawyers to accident sites? How soon until OnStar sells accident records so used-car purchasers can learn the vehicle's history? Chris Norloff ------------------------------ Date: Wed, 6 Jun 2001 12:17:49 -0400 From: Declan McCullagh Subject: U.K. plans mandatory IP indoctrination for children (from Cluebot) http://www.cluebot.com/article.pl?sid=01/06/05/2338246 U.K. Plans Mandatory IP Indoctrination for Children posted by vergil on Wednesday June 06, @12:10PM from the get-em-while-they're-young dept. Forget digital watermarks and cease-and-desist letters. The future of intellectual property enforcement lies not in technological access controls or litigation, but mandatory education. Anthony Murphy, the UK Patent Office's Director of Copyright since 1999, has hit upon a novel solution to stamp out public disregard for copyright law by nipping future file-swappers in the bud. In a move that's an eerie cross between Brave New World and the Lehman Working Group's "Just Say Yes" (to licensing) proposal, the UK's Patent Office and Department of Education have teamed up to teach youngsters the virtues of copyright. Starting in fall 2002, reverence to intellectual property -- and, presumably, disdain for Napster and its successors -- will become part the "Citizenship" aspect of England's National Curriculum for secondary school students. According to a April 26, 2001 UK Patent Office press release: "In Autumn 2002, a new subject, Citizenship, is being introduced into the National Curriculum in UK secondary schools. Its aim is to teach children how to be good, moral, citizens and Anthony Murphy believes the subject would be an ideal vehicle for teaching children about intellectual property. 'By bringing awareness of the importance of copyright into our schools, tomorrow's consumers can take their place in a community which understands, values and respects intellectual property.'" POLITECH -- Declan McCullagh's politics and technology mailing list You may redistribute this message freely if you include this notice. To subscribe, visit http://www.politechbot.com/info/subscribe.html This message is archived at http://www.politechbot.com/ ------------------------------ Date: Thu, 07 Jun 2001 09:02:06 -0700 From: "Robert Meineke" Subject: Re: Billboard error message (PGN, RISKS-21.45) Just for fun, check out http://www.daimyo.org/bsod/ [This Web site shows some classic blue screens of death in very conspicuous places. PGN] ------------------------------ Date: Thu, 7 Jun 2001 11:15:50 -0700 From: Rick Prelinger Subject: Re: Billboard error messages (PGN, RISKS-21.45) The best CalTrans error message I have seen was sometime last fall on the San Francisco approach to the Golden Gate Bridge, where an industrious purple LED sign repeatedly flashed "NO DATA." Rick Prelinger, Prelinger Archives, P.O. Box 590622, San Francisco, Calif. 94159-0622 +1 415 750-0445 http://www.prelinger.com footage@panix.com ------------------------------ Date: Fri, 8 Jun 2001 00:16 +0100 (BST) From: jgd@cix.co.uk (John Dallman) Subject: Re: Billboard error message (PGN, RISKS-21.45) My personal favourite was the time I found a hole-in-the-wall cash dispenser that had fallen over and was displaying a "C:>" prompt. A little playing with the keyboard revealed that MS-DOS was running - or something else that said "Bad command or file name" - and the keypad gave me numbers, ESC, BACKSPACE and ENTER. With no ALT key or letters, I couldn't do more, so the design had some limited degree of fail-safety. John Dallman ------------------------------ Date: Mon, 4 Jun 2001 21:57:48 -0400 From: Jamie McCarthy Subject: Re: Risks of clueless marketing (Searle, RISKS-21.44) > Has anybody else realized that "XP" is a person wincing [...]? This is the company that named an earlier operating system "WinCE". Maybe their *market* is people with pained facial expressions. ------------------------------ Date: Thu, 07 Jun 2001 16:46:45 -0400 From: Mike Nuss Subject: Re: Steve Gibson: Windows XP Vulnerable; Big ISPs just don't care I felt I had to respond to this article, because it's simply ridiculous. Raw sockets support, the supposed "vulnerability," is not a security risk. This capability is already present in every major Unix operating system, and can be acquired in every version of Windows with the addition of a library. >From atstake.com: The "powerful Internet-connection capabilities" which are hyped in this article is merely the ability to write raw IP packets. This is where an application program controls every field in the IP packet. This functionality is required if you were writing your own network bridge program for Windows or other low level network applications. An IDS for NT that resets connections would need this functionality. AntiSniff, which detects sniffers on a network, requires this functionality. This capability, which this article states is so dangerous to the Internet, is already available practically everywhere. It is available in every commercial and open source unix distribution and is already available for all Windows platforms (not just Windows XP) through the use of free add on libraries such as winpcap and libnetNT. The hype and hyperbole is astounding. From reading this article you'd think a deluge of DDoS attacks was building up just waiting to be released once Microsoft releases the all powerful new API. Nothing could be further from the truth. When XP arrives it will receive a collective yawn from DDoS attackers who would much rather have their win32 DDoS clients run on every version of Windows using the already available add on libraries. Once an attacker has administrative control of a machine they can run any code they want, whether it is native or in an uploaded executable. There is absolutely nothing stopping an attacker from spoofing IP addresses from a Windows machine today or tomorrow. The real RISK here is *The New York Times'* propagation of false information for the sole purpose of provoking Fear, Uncertainty, and Doubt. Mike Nuss ------------------------------ Date: Thu, 07 Jun 2001 00:48:25 -0500 From: David Crooke Subject: Re: Steve Gibson's report and Windows XP "Vulnerabilities" I have to take issue with Steve's assessment of how important this new capability in Windows 2000 / XP is - given the technical mastery required to subvert a machine in the first place, it's not a major endeavour to implement one's own source IP spoofing in any number of ways - a second virtual interface, bundling a custom IP stack with the trojan, or just changing the IP address of the machine. The fact that most current attacks don't use IP spoofing is not because Microsoft has failed to provide a convenient API - attackers simply haven't felt the need. Other operating systems have "supported" IP spoofing for years without it being regarded as risk contributing to hacking efforts. The real takeaway from Steve's write-up is that the endpoints of the Internet can no longer be trusted; it is time for network administrators at ISPs, universities and commercial premises to take up the cudgel and police the traffic emanating from their networks; source IP filtering is trivial to implement at this level. It is also time for backbone providers to introduce sensible firebreaks and reduce their trust in traffic passing through their systems. ------------------------------ Date: Thu, 7 Jun 2001 12:55:52 -0700 From: Stef Maruch Subject: They're at it again: Internet Explorer Smart Tags in WinXP A while back, when www.deja.com still archived Usenet news, they tried to generate revenue by inserting URLs into Usenet posts archived on their site. Needless to say, this upset a lot of Usenet posters, who considered it a copyright violation. Now Microsoft is up to much the same thing with a new feature of WinXP called "Internet Explorer Smart Tags": http://public.wsj.com/sn/y/SB991862595554629527.html In effect, Microsoft will be able, through the browser, to re-edit anybody's site, without the owner's knowledge or permission, in a way that tempts users to leave and go to a Microsoft-chosen site -- whether or not that site offers better information. Seems to me they should be called "Internet Explorer Sneak Tags." Stef ** rational/scientific/philosophical/mystical/magical/kitty ** stef@cat-and-dragon.com <*> http://www.cat-and-dragon.com/~stef ** I mean, 'e' was *already* the most common letter in the English language. -- AM, complaining about the online commerce explosion ------------------------------ Date: Thu, 7 Jun 2001 18:41:07 +1000 From: Andy Newman Subject: Re: Office XP modifies what you type (Deegan/Arnold, RISKS-21.42) When I saw the headline I thought "Oh, oh, MS at it again" but after reading further on must agree with what they're doing. A quick glance at an appropriate RFC - 2396, Uniform Resource Identifiers: Generic Syntax - shows that forward slash is reserved within URI paths and may not appear twice in succession. I quote, The path may consist of a sequence of path segments separated by a single slash "/" character. Within a path segment, the characters "/", ";", "=", and "?" are reserved. Also having written a few simple web servers and many robots I find the claim that there are many uses of '//' rather dubious. The people are probably thinking that some kind server's path normalisation is normal or the laziness of many HTTP server authors in transforming "entity" paths into the names of files storing those entities makes their invalid URLs allowable. I think the real risk of URLs (and I's and N's) is that they appear too similar to the names used in many file systems. This leads to things like thinking '//' in the middle of a path is valid (hey Unix copes!) or that ".jpg" on the end of a URL actually means something and you can ignore the entity type sent back with the data (common browser problem). Andy Newman, Silverbrook Research, ------------------------------ Date: Thu, 7 Jun 2001 15:24:18 -0400 From: "Jennings, Jay" Subject: Re: Office XP modifies what you type (Deegan/Arnold, RISKS-21.42) Two interesting points. First, in previous versions of Microsoft Word, the feature that changed capital letters could be turned off - it was called the "Auto Correct" feature and could be tweaked through the tools menu. The second point is more ironic. I received the link below in an e-mail yesterday: http://shop.microsoft.com//Products/Products_Feed/Online/SQLServer2000%5B101 45%5D/ProductQuestions.asp I was quickly able to deduce that Office XP was not used to compose the e-mail. Jay Jennings ------------------------------ Date: Thu, 7 Jun 2001 07:39:45 -0700 From: "Ratcliffe, Jackson" Subject: Microsoft, 'Mitigating Factors' and Public Relations Microsoft recently announced yet another security flaw, this one related to Exchange 2000's Outlook Web Access (OWA). Apparently java/vbscript attachments are automatically run http://www.microsoft.com/technet/security/bulletin/MS01-030.asp with no security. This is a REAL glaring flaw. So to make sure that it doesn't sound quite so bad, in Microsoft's e-mail announcement they tried to list the mitigating factors. Have a laugh. Mitigating Factors: - The vulnerability could only be exploited if the user were using OWA in conjunction with IE. (isn't that the whole point of the product ?) - The vulnerability is only exploitable by attachments that are received via OWA. In general, an attacker would have no way to determine whether a user would open an attachment using OWA rather than an Outlook client. (Isn't the whole point of .net to get rid of client-based Outlook?) [CC:ed on this item by Jackson, Gregory D. Marx concludes that "based on the first mitigating factor, I guess MS is suggesting that we switch to Netscape!?!?!" PGN] ------------------------------ Date: Wed, 6 Jun 2001 22:56:34 -0700 From: "Steve Loughran" Subject: Broken shopping carts I was just trying to by something from an on-line catalog (autosport.com), but was having problems as the shopping cart doubled the number of items I entered; the minimum purchase was two. On a whim, I entered a negative number -and the shopping cart updated to show that I was ordering -2 items, and had to pay -$188. I didn't go ahead with the transaction, but it would be an interesting experiment to see whether it would actually be possible to get free cash from shopping at this web site. It would also be interesting to see if the credit card companies fraud protection works in reverse -detecting and flagging too many refunds coming from a single vendor. ------------------------------ Date: Mon, 4 Jun 2001 16:07:56 +0100 From: Clive Page Subject: How to avoid Internet interruption at AAS meeting Astronomers planning to attend the American Astronomical Society meeting on now were advised as follows in an e-mail circular: If you plan on attending the AAS Meeting in Pasadena, CA 3-7 June 2001, you will most likely want to use the Meeting's Cyber Cafe for E-mail and Web Browsing. In order to ensure continuous access to your home site, please notify your local system and security administrators of the following: The Internet traffic flowing from the meeting attendees, will be coming from the IP addresses ranging from [CENSORED... actual addresses removed for obvious reasons]. In the past government sites have become aware of heavy traffic from our meetings and without notice shut off ALL access to attendees. This was done as a security measure, unaware that the traffic was originating at an AAS Annual Meeting. It caused several days of service interruption for meeting registrants. Informing your system administrators of the IP addresses could save you a lot of distress later! The risk: trying to avoid denial-of-service attacks might cause almost as much disruption to your staff as an actual attack, and just when they are least likely to be able to do much about it. Clive Page, Dept of Physics & Astronomy, University of Leicester. U.K. ------------------------------ Date: Tue, 05 Jun 2001 10:12:33 -0700 From: Fred Gilham Subject: There's no such thing as software `piracy' I know it's not a new idea, but I think it needs to be reiterated that piracy (which apparently is still practiced in some parts of the world) is a crime of violence, often resulting in the death of its victims, whereas making unauthorized copies of software that is copyright or licensed, while illegal in most places, is not a crime of violence. It may be tilting at windmills, like trying to get people to use the term `crackers' instead of `hackers'. Perhaps the people who write stories about this stuff would be more careful with their terminology if people started referring to `taggers' (i.e., graffiti vandals) as `journalists'? After all, they both work with words.... ------------------------------ Date: 31 May 2001 10:18:07 -0400 From: huggins@quip.eecs.umich.edu (James K. Huggins) Subject: Re: Another fear of Risks Sorry ... here I go on a rant ... "Bob Frankston" writes: > I'm using IE 6.0 and it works pretty much like 5.0. With one notable > exception -- UPS explicitly checks for it and doesn't let me use their > service with an unapproved browser. I presume that feel it is better for > them to lose customers than risk .. risk what? Risk spending countless hours of time on the phone (and therefore $$) with irate customers blaming UPS when the customers' new-fangled "compatible" browser doesn't work with the UPS site. Risk having people blame UPS instead of Microsoft when IE 6.0 turns out to not be 100% compatible with IE 5.x in a couple of features which the UPS cite depends upon to function correctly ... especially if those incompatibilities didn't surface in any of the pre-release versions. > UPS is loses two ways. They force me to use other services and they > lose the value of users doing testing for them. In my humble opinion, most users aren't interested in doing testing for companies. That's what we pay the companies to do for themselves. Furthermore, relying on user reports for testing is full of its own problems. Users (and I count myself in that category) will often blame others for problems they cause themselves, or problems caused by third parties (e.g. ISPs) which aren't the fault of either endpoint. > They can warn me that they haven't tested with my browser but > disallowing it is not only short-sighted, it represents a basic > misunderstanding of the PC and the large effort put in to assure > compatibility with previous versions of programs. Who says UPS won't eventually support IE 6.0? Given that it's just been released, UPS may just be trying to give itself some time to test IE 6.0 for itself and fix any compatibility problems on its end. > Old MIS (before they were called IT) departments did have a great > fear of upgrades since each mainframe system was extensively > patched. But that reasonable fear is now a phobia. Nope. Look, I've had much the same problem with the Netscape 4->6 transition. When I upgraded to the "improved" Netscape 6 on my home machine, lots of sites that I used to visit simply refused to work anymore. When I contacted the sites to complain, most state that the problem is Netscape's and that I should either downgrade back to 4.72 or switch to IE. There ain't nothing that's 100% backward compatible, especially in a x.0 release. Just my $.02. --Jim Huggins, Kettering University, Flint, MI (jhuggins@kettering.edu) ------------------------------ Date: Tue, 29 May 2001 22:11:30 -0400 (EDT) From: Jeffrey Jonas Subject: Re: McDonald's testing cashless payments (RISKS-21.43) > McDonald's Corporation has begun testing the use of a cashless payment > system that uses the kind of radio transponder technology that was first > developed by state highways to allow motorists to drive through toll plazas > without having to stop to make a payment. A friend said that McD's once had a credit card but dropped it. Sure, it made checkouts faster and less handling of cash, but it had an unexpected side effect. Folks saw the monthly bill and realized how all those meals were adding up to real money and cut back their spending since it was so easily auditable. Another interesting interaction: > Newsgroups: alt.consumers.experiences,misc.consumers > Subject: Re: McDonald's 30-Second DT Guarantee McD's apparently has some promotion where they guarantee you get the food 30 seconds after paying. The immediate analysis is that they'll take as long as before, just not collect the money 'till it's ready. Now with the speed-pass, will the guarantee still hold? ------------------------------ Date: 30 May 2001 02:26:29 -0400 From: "John R Levine" Subject: Re: McDonald's testing cashless payments (RISKS-21.43) I had a Mobil speedpass for a while. It's about the diameter of a pencil and an inch long, with a hole through the end so it can go on your keychain. You wave it at the pump, a light on the pump goes on to tell you it knows who you are and you pump your gas. Mobil links theirs to a credit card. It worked fine until one day my bank called me up to say that I had been buying an awful lot of gas in towns east of here, had I lost my card? No, but it turned out that I'd lost my speedpass. It fell off my keychain the last time I used it, but it was so small that I didn't notice it was gone, what with all the frequent shopper barcode tags et al with my keys. I finally got it straightened out and Mobil ate the bogus charges, a relief since the card company said their usual anti-fraud rules don't apply when you don't use your physical card for a transaction. I decided I'll spend the extra two seconds per visit and swipe my card. I do have an E-ZPass toll transponder in my truck, but that's different for two reasons: it's large enough to miss and is firmly glued to the inside of the windshield, and they give me the incentive of significant toll discounts (in NYC at least) if I use it. John Levine, johnl@iecc.com, Primary Perpetrator of "The Internet for Dummies", Information Superhighwayman wanna-be, http://iecc.com/johnl, Sewer Commissioner ------------------------------ Date: Wed, 06 Jun 2001 19:55:27 -0500 From: William Paul Fiefer Subject: Credit where it isn't due So you request a credit card and it comes by mail with a peel-off sticker across the signature plate. The sticker tells you to call a toll-free number to activate the card. This is, apparently, a theft-prevention thing. Don't bother. The cards activate automatically. At least "Blue" from American Express and the "Platinum" series ($100,000 credit limit -- $250,000 for the "Quantum" series) from MBNA do. I ordered these cards but did not activate them. I found myself receiving mail regarding these accounts. I received privacy notices, which I opted out of. Then I asked MBNA why I had a card I did not activate. If you do not activate our cards, the customer rep said, they activate themselves after a set time limit. The American Express rep told me no such activation occurred but could not explain why my card was active. She even tried to discourage me from cancelling the thing! The RISK? You'll have credit due where none is applied for. William Paul Fiefer 630.892.5180 www.prairienet.org/~yamada ------------------------------ Date: 12 Feb 2001 (LAST-MODIFIED) From: RISKS-request@csl.sri.com Subject: Abridged info on RISKS (comp.risks) The RISKS Forum is a MODERATED digest. Its Usenet equivalent is comp.risks. => SUBSCRIPTIONS: PLEASE read RISKS as a newsgroup (comp.risks or equivalent) if possible and convenient for you. Alternatively, via majordomo, SEND DIRECT E-MAIL REQUESTS to with one-line, SUBSCRIBE (or UNSUBSCRIBE) which now requires confirmation to majordomo@CSL.sri.com (not to risks-owner) [with option of E-mail address if not the same as FROM: on the same line, which requires PGN's intervention -- to block spamming subscriptions, etc.] or INFO [for unabridged version of RISKS information] .MIL users should contact (Dennis Rears). .UK users should contact . => The INFO file (submissions, default disclaimers, archive sites, copyright policy, PRIVACY digests, etc.) is also obtainable from http://www.CSL.sri.com/risksinfo.html ftp://www.CSL.sri.com/pub/risks.info The full info file will appear now and then in future issues. *** All contributors are assumed to have read the full info file for guidelines. *** => SUBMISSIONS: to risks@CSL.sri.com with meaningful SUBJECT: line. => ARCHIVES are available: ftp://ftp.sri.com/risks or ftp ftp.sri.comlogin anonymous[YourNetAddress]cd risks [volume-summary issues are in risks-*.00] [back volumes have their own subdirectories, e.g., "cd 20" for volume 20] http://catless.ncl.ac.uk/Risks/VL.IS.html [i.e., VoLume, ISsue]. Lindsay Marshall 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/ . http://www.planetmirror.com/pub/risks/ ftp://ftp.planetmirror.com/pub/risks/ ==> 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 ------------------------------ End of RISKS-FORUM Digest 21.46 ************************