Long-torsoed anatomy model Michael Phelps has won more gold medals in one Olympics than anyone before him. That's not the only record he's set in the last week. According NBC's Bob Costas, Phelps has more Facebook "fans" than Will Smith, Miley Cyrus, and the Jonas Brothers — 767,885 at last count! Phelps tells Costas that besides the fans, he's got about 7,600 pending Facebook friend requests, too. "I can't accept any more," he tells Costas. Obsessed? You can always try. You might have better luck at friending Phelps than becoming his fan. Check out the screenshot below — that feature seems to be broken right now, perhaps because of the sudden onrush of Phelpsmania on Facebook.

A Tesla roadster dead at the intersection of Sand Hill Road and Alameda in Menlo Park, at approximately 4 p.m. yesterday. Yes, the sexy electric car that's the darling of Valley cleantech investors, on the road that runs through the heart of venture-capitalist country. With so few Teslas in the wild, there can't be that many possibilities on who the unlucky buyer was. Let us know in the comments which rich guy you think was late getting home last night. The big question is, was it a mechanical breakdown, or did the batteries run out of juice? While others may look at this and think Tesla can't manufacture a reliable car, we see this as an opportunity for the fledgling automotive startup: Electric tow trucks.

How did Netflix end up with massive delivery delays? "Because of massive database corruption in their Oracle cluster caused by a botched upgrade," according to a tipster. But don't blame Oracle (or Microsoft), necessarily. The tipster believes Netflix's own database adminstrators who bungled the upgrade. Why not just roll back the system?

Woe be unto Netflix if my parents don't get the latest installemnt of Foyle's War. In an email sent out to customers and a notice posted to the site, the DVD-by-mail company says it is having problems with its shipping system affecting around a third of the company's customers. It has now persisted for two days. So if your friendly mail carrier doesn't show up with a red envelope or three today, don't blame it on a Postal Service "blue shorts of death" error. Graciously, the company has preemptively offered a credit for any delays. Why not tout its online-video offerings, like Watch Now streaming on its website or the Roku set-top box? Oh, right, website outages and inventory problems. But hey, at least if your request gets returned "404 Not Found," it won't cost you a stamp. Netflix's alert, after the jump:

The website of the New York Times is unresponsive beyond an archaic "Http/1.1 Service Unavailable" error. We didn't even have to check the site — we could have gathered as much from the frantic IMs being sent by the same Internet poseurs who like to blog about how the mainstream media is irrelevant to their lives.

The enigmatic David G. of Apple has been given the go-ahead to proclaim MobileMe's email problems, affecting those lucky 1 percent of users, resolved after three weeks. I guess someone should email the FailMe Is More Like It guy. [Apple]

HD Moore is the guy behind the Metasploit Project. In general, Metasploit helps sysadmins find security holes in their networks. Last week, Moore published an exploit for a weakness in the domain name server (DNS) software used to route Web surfers to the correct machine for, say, www.valleywag.com. On Tuesday, some of the traffic to Moore's employer, BreakingPoint, was rerouted to a fake Google page operated by a scammer running a click fraud racket. The cause? An AT&T DNS server for the Austin, Texas area that had been compromised using ... you guessed it. Moore emailed us, "There is no way to verify now that it has been fixed, but my impression is that it was actually a different exploit."

LiveJournaler akil writes of a recent visit to the Apple Store, where a new, streamlined process for iPhone buying was in effect: "They started prequalifying people at 6:30 a.m. Within three minutes of arriving, I was given a serialized tag that is linked to an actual iPhone and I'm guaranteed to get one." Separately, an Apple employee who gives his name only as "David G." says Steve has asked him to post regularly on the status of Apple's buggy MobileMe service. (Photo by akil)

Apple's .Mac email — relaunched as MobileMe in conjunction with the iPhone 3G two Fridays ago — is still flying as crooked as Drinky Crow on payday. MacRumors has aggregated customer gripes. Apple's hard-to-swallow response: Only 1 percent of customers are having problems after Apple's server migration. MobileMe mail works for stationary old me, but see these screenshots from readers:

An unexplained failure briefly knocked out Amazon.com's Simple Storage Service this morning, taking with it parts of Twitter, The Huffington Post, and many other sites. You can track Amazon status yourself — see the red "Service disruption" icons under Status History. Has anyone yet built RescueTime for sites instead of people? We could park a Web 2.0 Failboard on our front page.

Twitter got its big break with mainstream America today — a big article in the free Ramada Inn daily, USA Today. In fact, the article drove so much traffic to Twitter that — whoops — it crashed. Again. "They used to call it the Slashdot Effect," Valleywag's resident Olds Paul Boutin tells me, "Your site would go down right in the middle of your moment of glory." Smartly, though, Twitter PR managed to get USA Today's Jefferson Graham to build an excuse for the site's Love For Fail right into the article. Graham reports:

NEW YORK — Apple's iTunes store, required for activating the new iPhone 3G is failing, causing massive chaos from coast to coast. Even Apple employees are — when they don't realize a reporter is in earshot — acknowledging this. "I can't believe there's just so much stuff going wrong," says one employee at the Fifth Avenue Apple Store as he takes his lunch break sitting next to me. "It's not very Apple-like. It's shitty. It just shouldn't happen." His friend agrees: "I called my dad and his phone still doesn't work."

NEW YORK — Apple Store employees are a little tense today. They got nine hours of training preparing for today's iPhone 3G launch. Then there was all the press and hoopla when the day finally began. (I overheard two of them complaining about it: "I felt like I was going to vomit," one said. The other: "I felt like was as going to vomit too!") Then there was the crowd control. Then the iTunes Store, required to activate phones and thereby complete sales, went down. I snuck a hidden camera into the Fifth Avenue Apple Store and surveyed the chaos. Roll the clip. Meanwhile, here's a reader's account of an experience at an Apple Store in Walnut Creek, California:

Sometimes a picture is worth 100 words. Note the "VA" up top, too. (For non-Valleys, Google Street View shows where Yahoo's famous roadside billboard sits along the San Francisco onramp to the Bay Bridge.) Yahoo's lavish parody of a Holiday Inn sign went up during the craziest part of the Web 1.0 boom. The message — "A nice place to stay on the Internet" — was crafted to counter the conventional wisdom that Web surfers used "portals" like Yahoo for a quick search, then clicked away to spend their time at "destination" sites somewhere else. Yahoo needed to convince investors and everyone else that its site was a destination, not a portal, and that it was "sticky" even though everyone eventually went somewhere else after a few minutes. Hence the Internet-motel metaphor. Pop quiz for geeks: Can someone calculate the monthly electric bill for this thing? (Photo by Ben Roodman)

British wireless carrier O2 saw its website fail after receiving 13,000 preorders per second for Apple's new iPhone 3G. Whatever happened to British reserve? [Digital Daily]

Google Docs, one of the permanent-beta applications that Google offers as a free, lightweight alternative to Microsoft Office, went down for a brief spell just before 9 a.m. While it's back up now, as a reader pointed out, it's "Not very convenient when you have important files you need for a meeting." Welcome to the future of cloud computing, where instead of a blue screen of death, you get a server error message.

Tech workers looking for cash are selling information about vulnerabilities in their own companies' products, according to a report in Fast Company by investigative journalist Adam Penenberg. (For the Olds, Penenberg is the guy who busted hacker-hoax writer Stephen Glass ten years ago. Yes, ten years. We are OLDZ.) Penenberg got Hewlett-Packard to admit they'd been compromised by "a rogue employee in France," then tracked down the guy he believes bought the info: An instructor at Paris's Institut Supérieur d'Electronique.

Plagued by breakdowns — and by Robert Scoble — Twitter has hired Pivotal Labs to help it fix the messaging service. "Yes, we hired some smart folks from Pivotal Labs to join us in the office and get some work done," Twitter cofounder Biz Stone told VentureBeat. "Joining us in the office" sounds impressive, but it involves a five-block commute — Pivotal Labs, a software consultancy, is part of the insular San Francisco Web-startup scene that spawned Twitter in the first place. The Pivotal crew are old hands indeed at taking money from the Internet-flummoxed likes of Stone. How do I know this? Full disclosure: Pivotal executive Ian McFarland helped me launch a website — in 1995.

After a considerable outage that started around 10:30 a.m. Pacific Time, Amazon.com's homepage is now back up and still trying to get someone, anyone, to buy a Kindle. Shortly after the outage, the company issued a short statement in the forums for 3rd party sellers titled "Amazon Un-Planned Event" — not such an unusual occurrence, apparently, but rarely of this magnitude:

It's one thing when Twitter goes limp (again). It's another when e-commerce oligarch Amazon.com can't keep it up for customers. But that's just what readers are reporting, and we've confirmed. I mean, if you're going to rule the Web, it really is best to remain virile atop your throne. Sure it means lost sales for Amazon — and for its network of affiliates — but our question is, are Web startups using the company's "cloud" services such as EC2 also experiencing problems? Update: Even an attempt to access the site via IP address instead of domain name is for naught, ruling out a simple DNS issue.