As an early domain name investor, Mike O’Connor had by 1994 snatched up several choice online destinations, including bar.com, cafes.com, grill.com, place.com, pub.com and television.com. Some he sold over the years, but for the past 26 years O’Connor refused to auction perhaps the most sensitive domain in his stable — corp.com. It is sensitive because years of testing shows whoever wields it would have access to an unending stream of passwords, email and other proprietary data belonging to hundreds of thousands of systems at major companies around the globe.

Now, facing 70 and seeking to simplify his estate, O’Connor is finally selling corp.com. The asking price — $1.7 million — is hardly outlandish for a 4-letter domain with such strong commercial appeal. O’Connor said he hopes Microsoft Corp. will buy it, but fears they won’t and instead it will get snatched up by someone working with organized cybercriminals or state-funded hacking groups bent on undermining the interests of Western corporations.

One reason O’Connor hopes Microsoft will buy it is that by virtue of the unique way that Windows handles resolving domain names on a local network, virtually all of the computers trying to share sensitive data with corp.com are somewhat confused Windows PCs. More importantly, early versions of Windows actually encouraged the adoption of insecure settings that made it more likely Windows computers might try to share sensitive data with corp.com.

At issue is a problem known as “namespace collision,” a situation where domain names intended to be used exclusively on an internal company network end up overlapping with domains that can resolve normally on the open Internet.

Windows computers on an internal corporate network validate other things on that network using a Microsoft innovation called Active Directory, which is the umbrella term for a broad range of identity-related services in Windows environments. A core part of the way these things find each other involves a Windows feature called “DNS name devolution,” which is a kind of network shorthand that makes it easier to find other computers or servers without having to specify a full, legitimate domain name for those resources.

For instance, if a company runs an internal network with the name internalnetwork.example.com, and an employee on that network wishes to access a shared drive called “drive1,” there’s no need to type “drive1.internalnetwork.example.com” into Windows Explorer; typing “\\drive1\” alone will suffice, and Windows takes care of the rest.

But things can get far trickier with an internal Windows domain that does not map back to a second-level domain the organization actually owns and controls. And unfortunately, in early versions of Windows that supported Active Directory — Windows 2000 Server, for example — the default or example Active Directory path was given as “corp,” and many companies apparently adopted this setting without modifying it to include a domain they controlled.

Compounding things further, some companies then went on to build (and/or assimilate) vast networks of networks on top of this erroneous setting.

Now, none of this was much of a security concern back in the day when it was impractical for employees to lug their bulky desktop computers and monitors outside of the corporate network. But what happens when an employee working at a company with an Active Directory network path called “corp” takes a company laptop to the local Starbucks?

Chances are good that at least some resources on the employee’s laptop will still try to access that internal “corp” domain. And because of the way DNS name devolution works on Windows, that company laptop online via the Starbucks wireless connection is likely to then seek those same resources at “corp.com.”

In practical terms, this means that whoever controls corp.com can passively intercept private communications from hundreds of thousands of computers that end up being taken outside of a corporate environment which uses this “corp” designation for its Active Directory domain.

INSTANT CORPORATE BOTNET, ANYONE?

That’s according to Jeff Schmidt, a security expert who conducted a lengthy study on DNS namespace collisions funded in part by grants from the U.S. Department of Homeland Security. As part of that analysis, Schmidt convinced O’Connor to hold off selling corp.com so he and others could better understand and document the volume and types of traffic flowing to it each day.

During an eight month analysis of wayward internal corporate traffic destined for corp.com in 2019, Schmidt found more than 375,000 Windows PCs were trying to send this domain information it had no business receiving — including attempts to log in to internal corporate networks and access specific file shares on those networks.

For a brief period during that testing, Schmidt’s company JAS Global Advisors accepted connections at corp.com that mimicked the way local Windows networks handle logins and file-sharing attempts.

“It was terrifying,” Schmidt said. “We discontinued the experiment after 15 minutes and destroyed the data. A well-known offensive tester that consulted with JAS on this remarked that during the experiment it was ‘raining credentials’ and that he’d never seen anything like it.”

Likewise, JAS temporarily configured corp.com to accept incoming email.

“After about an hour we received in excess of 12 million emails and discontinued the experiment,” Schmidt said. “While the vast majority of the emails were of an automated nature, we found some of the emails to be sensitive and thus destroyed the entire corpus without further analysis.”

Schmidt said he and others concluded that whoever ends up controlling corp.com could have an instant botnet of well-connected enterprise machines.

“Hundreds of thousands of machines directly exploitable and countless more exploitable via lateral movement once in the enterprise,” he said. “Want an instant foothold into about 30 of the world’s largest companies according to the Forbes Global 2000? Control corp.com.”

THE EARLY ADVENTURES OF CORP.COM

Schmidt’s findings closely mirror what O’Connor discovered in the few years corp.com was live on the Internet after he initially registered it back in 1994. O’Connor said early versions of a now-defunct Web site building tool called Microsoft FrontPage suggested corporation.com (another domain registered early on by O’Connor) as an example domain in its setup wizard.

That experience, portions of which are still indexed by the indispensable Internet Archive, saw O’Connor briefly redirecting queries for the domain to the Web site of a local adult sex toy shop as a joke. He soon got angry emails from confused people who’d also CC’d Microsoft co-founder Bill Gates.

O’Connor said he also briefly enabled an email server on corp.com, mainly out of morbid curiosity to see what would happen next.

“Right away I started getting sensitive emails, including pre-releases of corporate financial filings with The U.S. Securities and Exchange Commission, human resources reports and all kinds of scary things,” O’Connor recalled in an interview with KrebsOnSecurity. “For a while, I would try to correspond back to corporations that were making these mistakes, but most of them didn’t know what to do with that. So I finally just turned it off.”

TOXIC WASTE CLEANUP IS HARD

Microsoft declined to answer specific questions in response to Schmidt’s findings on the wayward corp.com traffic. But a spokesperson for the company shared a written statement acknowledging that “we sometimes reference ‘corp’ as a label in our naming documentation.”

“We recommend customers own second level domains to prevent being routed to the internet,” the statement reads, linking to this Microsoft Technet article on best practices for setting up domains in Active Directory.

Over the years, Microsoft has shipped several software updates to help decrease the likelihood of namespace collisions that could create a security problem for companies that still rely on Active Directory domains that do not map to a domain they control.

But both O’Connor and Schmidt say hardly any vulnerable organizations have deployed these fixes for two reasons. First, doing so requires the organization to take down its entire Active Directory network simultaneously for some period of time. Second, according to Microsoft applying the patch(es) will likely break or at least slow down a number of applications that the affected organization relies upon for day-to-day operations.

Faced with either or both of these scenarios, most affected companies probably decided the actual risk of not applying these updates was comparatively low, O’Connor said.

“The problem is that when you read the instructions for doing the repair, you realize that what they’re saying is, ‘Okay Megacorp, in order to apply this patch and for everything to work right, you have to take down all of your Active Directory services network-wide, and when you bring them back up after you applied the patch, a lot of your servers may not work properly’,” O’Connor said.

Curiously, Schmidt shared slides from a report submitted to a working group on namespace collisions suggesting that at least some of the queries corp.com received while he was monitoring it may have come from Microsoft’s own internal networks.

“The reason I believe this is Microsoft’s issue to solve is that someone that followed Microsoft’s recommendations when establishing an active directory several years back now has a problem,” Schmidt said.

“Even if all patches are applied and updated to Windows 10,” he continued. “And the problem will persist while there are active directories named ‘corp’ – which is forever. More practically, if corp.com falls into bad hands, the impact will be on Microsoft enterprise clients – and at large scale – paying, Microsoft clients they should protect.”

Asked why he didn’t just give corp.com to Microsoft as an altruistic gesture, O’Connor said the software giant ought to be accountable for its products and mistakes.

“It seems to me that Microsoft should stand up and shoulder the burden of the mistake they made,” he said. “But they’ve shown no real interest in doing that, and so I’ve shown no interest in giving it to them. I don’t really need the money. I’m basically auctioning off a chemical waste dump because I don’t want to pass it on to my kids and burden them with it. My frustration here is the good guys don’t care and the bad guys probably don’t know about it. But I expect the bad guys would like it.”

Further reading:

Mitigating the Risk of DNS Namespace Collisions (PDF)

DEFCON 21 – DNS May Be Hazardous to your Health (Robert Stucke)

Mitigating the Risk of Name Collision-Based Man-in-the-Middle Attacks (PDF)

Many modern vehicles let owners use the Internet or a mobile device to control the car’s locks, track location and performance data, and start the engine. But who exactly owns that control is not always clear when these smart cars are sold or leased anew. Here’s the story of one former electric vehicle owner who discovered he could still gain remote, online access to his old automobile years after his lease ended.

Mathew Marulla began leasing a Ford Focus electric vehicle in 2013, but turned the car back in to Ford at the end of his lease in 2016. So Marulla was surprised when he recently received an email from Ford.com stating that the clock in his car was set incorrectly.

Out of curiosity, Marulla decided to check if his old MyFordMobile.com credentials from 2016 still worked. They did, and Marulla was presented with an online dashboard showing the current location of his old ride and its mileage statistics.

The dashboard also allowed him to remotely start the vehicle, as well as lock and unlock its doors.

“It was a three-year lease from Ford and I turned it in to Ford four years ago, so Ford definitely knows I am no longer the owner,” Marulla said, noting that the dashboard also included historic records showing where the Focus had been driven in days prior.

“I can track its movements, see where it plugs in,” he said. “Now I know where the current owner likely lives, and if I watch it tomorrow I can probably figure out where he works. I have not been the owner of this vehicle for four years, Ford knows this, yet they took no action whatsoever to remove me as the owner in this application.”

Asked to comment on Marulla’s experience, a spokesperson for Ford said all Ford dealerships are supposed to perform a “master reset” as part of their used car checklist prior to the resale of a vehicle. A master reset (carried out via the vehicle’s SYNC infotainment screen by a customer or dealer) disassociates the vehicle from all current accounts.

“A master reset cleans phone data and removes previous Ford Pass and My Ford Mobile connections,” the company said in a statement released to KrebsOnSecurity. “Once complete, a previous owner will no longer be able to connect to the vehicle when they log in to My Ford Mobile or Ford Pass.”

As Marulla’s experience shows, if you’re in the market for a used car you should probably check whether it’s possible to reset the previous owner’s control and/or information before purchasing it, or at least ask the dealership to help you ensure this gets done once the purchase is made.

And if you’re thinking of selling your car, it’s a good idea to clear your personal data from the vehicle first. As the U.S. Federal Trade Commission advises, some cars have a factory reset option that will return the settings and data to their original state.

“But even after a factory reset, you may still have work to do,” reads an FTC consumer privacy notice from 2018. “For example, your old car may still be connected to subscription services like satellite radio, mobile Wi-Fi hotspots, and data services. You need to cancel these services or have them transferred to your new vehicle.”

By the way, this issue of de-provisioning is something of a sticky wicket, and it potentially extends well beyond vehicles to a number of other “smart” devices that end up being resold or refurbished. This is doubly so for Internet-connected/capable devices whose design may give the previous owner a modicum of access to or control over the device in question regardless of what steps the new owner takes to limit such access (particularly some types of security cameras).

A Pennsylvania man who operated one of the Internet’s longest-running online attack-for-hire or “booter” services was sentenced to five years probation today. While the young man’s punishment was heavily tempered by his current poor health, the defendant’s dietary choices may have contributed to both his capture and the lenient sentencing: Investigators say the onetime booter boss’s identity became clear after he ordered a bacon and chicken pizza delivered to his home using the same email address he originally used to register his criminal attack service.

David Bukoski, 24, of Hanover Township, Pa., pleaded guilty to running Quantum Stresser, an attack-for-hire business — also known as a “booter” or “stresser” service — that helped paying customers launch tens of thousands of digital sieges capable of knocking Web sites and entire network providers offline.

Investigators say Bukoski’s booter service was among the longest running services targeted by the FBI, operating since at least 2012. The government says Quantum Stresser had more than 80,000 customer subscriptions, and that during 2018 the service was used to conduct approximately 50,000 actual or attempted attacks targeting people and networks worldwide.

The Quantum Stresser Web site — quantumstress[.]net — was among 15 booter services that were seized by U.S. and international authorities in December 2018 as part of a coordinated takedown targeting attack-for-hire services.

Federal prosecutors in Alaska said search warrants served on the email accounts Bukoski used in conjunction with Quantum Stresser revealed that he was banned from several companies he used to advertise and accept payments for the booter service.

The government’s sentencing memorandum says Bukoski’s replies demanding to know the reasons for the suspensions were instrumental in discovering his real name.  FBI agents were able to zero in on Bukoski’s real-life location after a review of his email account showed a receipt from May 2018 in which he’d gone online and ordered a handmade pan pizza to be delivered to his home address.

While getting busted on account of ordering a pizza online might sound like a bone-headed or rookie mistake for a cybercriminal, it is hardly unprecedented. In 2012 KrebsOnSecurity wrote about the plight of Yuriy “Jtk” Konovalenko, a then 30-year-old Ukrainian man who was rounded up as part of an international crackdown on an organized crime gang that used the ZeuS malware to steal tens of millions of dollars from companies and consumers. In that case, Konovalenko ultimately unmasked himself because he used his Internet connection to order the delivery of a “Veggie Roma” pizza to his apartment in the United Kingdom.

Interestingly, the feds say their examination of Bukoski’s Internet browsing records showed he knew full well that running a booter service was punishable under federal law (despite disclaimers published on Quantum Stresser stating that the site’s owners weren’t responsible for how clients used the service).

“The defendant’s web browsing history was significant to investigators for a number of reasons, including the fact that it shows that the defendant browsed an article written by a prominent security researcher referencing both the defendant’s enterprise along with a competing service, including a link provided by the researcher in the article to an advisory posted by the FBI warning that the operation of booter services was potentially punishable under federal law,” reads the sentencing memo from Assistant U.S. Attorney Adam Alexander.

That’s interesting because the article in question was actually a 2017 KrebsOnSecurity story about a mobile app tied to a competing booter service that happened to share some of the same content as Quantum Stresser.

That 2017 story referenced an FBI advisory that had just been issued warning the use of booter services is punishable under the Computer Fraud and Abuse Act, and may result in arrest and criminal prosecution.

Bukoski was sentenced to five years of probation and six months of “community confinement.” The government suggested a lenient sentence considering the defendant’s ongoing health complications, which include liver failure.

On Sept. 11, 2019, two security experts at a company that had been hired by the state of Iowa to test the physical and network security of its judicial system were arrested while probing the security of an Iowa county courthouse, jailed in orange jumpsuits, charged with burglary, and held on $100,000 bail. On Thursday Jan. 30, prosecutors in Iowa announced they had dropped the criminal charges. The news came while KrebsOnSecurity was conducting a video interview with the two accused (featured below).

Gary DeMercurio, 43 of Seattle, and Justin Wynn, 29 of Naples, Fla., are both professional penetration testers employed by Coalfire Labs, a security firm based in Westminster, Colo. Iowa’s State Court Administration had hired the company to test the security of its judicial buildings.

Under the terms of their contract (PDF), DeMercurio and Wynn were permitted to impersonate staff and contractors, provide false pretenses to gain physical access to facilities, “tailgate” employees into buildings, and access restricted areas of those facilities. The contract said the men could not attempt to subvert alarm systems, force-open doors, or access areas that require protective equipment.

When the duo’s early-morning Sept. 11 test of the security at the courthouse in Dallas County, Iowa set off an audible security alarm, they followed procedure and waited on-site for the police. DeMercurio and Wynn said when the county’s sheriff deputies arrived on the scene just a few minutes later, they told the officers who they were and why they were there, and that they’d obtained entry to the premises via an unlocked door.

“They said they found a courthouse door unlocked, so they closed it from the outside and let it lock,” Dan Goodin of Ars Technica wrote of the ordeal in November. “Then they slipped a plastic cutting board through a crack in the door and manipulated its locking mechanism. (Pentesters frequently use makeshift or self-created tools in their craft to flip latches, trigger motion-detected mechanisms, and test other security systems.) The deputies seemed impressed.”

To assuage concerns they might be burglars, DeMercurio and Wynn produced an authorization letter detailing the job they’d been hired to do and listing the names and mobile phone numbers of Iowa state employees who could verify their story.

After contacting some of the court officials listed in the letter, the deputies seemed satisfied that the men weren’t thieves. That is, until Dallas County Sheriff Chad Leonard showed up.

“The pentesters had already said they used a tool to open the front door,” Goodin recounted. “Leonard took that to mean the men had violated the restriction against forcing doors open. Leonard also said the men attempted to turn off the alarm—something Coalfire officials vehemently deny. In Leonard’s mind that was a second violation. Another reason for doubt: one of the people listed as a contact on the get-out-of-jail-free letter didn’t answer the deputies’ calls, while another said he didn’t believe the men had permission to conduct physical intrusions.”

DeMercurio and Wynn were arrested, jailed, and held for nearly 24 hours before being released on a $100,000 bail. Initially they were charged with felony third-degree burglary and possessing burglary tools, although those charges were later downgraded to misdemeanor trespass.

What initially seemed to Coalfire as a momentary lapse of judgment by Iowa authorities quickly morphed into the surreal when state lawmakers held hearings questioning why and how someone in the state’s employ could have so recklessly endangered the safety and security of its citizens.

Judicial Branch officials in Dallas County said in response to this grilling that they didn’t expect Coalfire’s physical penetration testing to be conducted outside of business hours. State Sen. Amy Sinclair was quoted as telling her colleagues that “the hiring of an outside company to break into the courthouses in September created ‘significant danger, not only to the contractors, but to local law enforcement, and members of the public.'”

“Essentially a branch of government has contracted with a company to commit crimes, and that’s very troubling,” lamented Iowa state Sen. Zach Whiting. “I want to find out who needs to be held accountable for this and how we can do that.”

Those strong words clashed with a joint statement released Thursday by Coalfire and Dallas County Attorney Charles Sinnard:

“Ultimately, the long-term interests of justice and protection of the public are not best served by continued prosecution of the trespass charges,” the statement reads. “Those interests are best served by all the parties working together to ensure that there is clear communication on the actions to be taken to secure the sensitive information maintained by the judicial branch, without endangering the life or property of the citizens of Iowa, law enforcement or the persons carrying out the testing.

Matthew Linholm, an attorney representing DeMercurio and Wynn in the case, said the justice system ceases to serve its crucial function and loses credibility when criminal accusations are used to advance personal or political agendas.

“Such a practice endangers the effective administration of justice and our confidence in the criminal justice system,” Linholm told The Des Moines Register, which broke the news of the dropped charges.

While the case against Coalfire’s employees has rallied many in the cybersecurity community around the accused, not everyone sees this dispute in black-and-white. Chris Nickerson, a digital intrusion specialist and founder of LARES Consulting, said in a Twitter post Thursday that “when a company puts us in harm’s way due to their poor planning, failed sales education, inadequate project management and deplorable contract management…We shouldn’t celebrate them. We should hold them accountable.”

Asked to elaborate, Nickerson referred to a recent podcast which touched on the arrests.

“The things that concern me about this situation are more of the pieces of safety that exist across how the industry instruments doing these types of engagements,” Nickerson said. “They seem very, very reasonable and obvious once they become obvious but until then they’re completely foreign to people.”

“It’s really on the owners of the organization to educate the customer of those potential pitfalls,” Nickerson continued. “Because there isn’t a good standard. We haven’t all gotten together and institutionalized the knowledge that we have in our heads and dump it down to paper so that someone who is new to the field being tasked with this can go through and say, ‘Hey, did you ask them if the city versus the state versus the building owner and the real estate people…are all of these people in lock step?'”

Coalfire CEO Tom McAndrew seemed to address this point in our interview Thursday, saying there were two unique aspects of this particular engagement. First, although the client in this case said they did not want Coalfire to make local law enforcement aware of the ongoing engagement prior to testing the physical security of the site, it was clear after the fact that state officials never did that on their own.

More importantly, McAndrew said, there was ambiguity around who actually owned the buildings that they were hired to test.

“If you’re doing a test for the state and you walk into the building and it’s the courthouse and you’re doing a test for the court system, you’d think that they would have jurisdiction or own it, and that turned out not to be the case in this scenario because there’s some things the state owns and some things the county owns, and that was something we weren’t aware of as we did some of this work,” he said. “We didn’t understand the nuances.”

Asked what Coalfire has learned from this ordeal, McAndrew said his company is likely to insist that local, state and even federal law enforcement be informed in advance of any penetration tests, at least as far as those engagements relate to public entities.

“When we look at the contracts and we look at who’s authorized to do what…typically, if a [chief security officer] says test these IP addresses, we would say okay that’s enough,” he said. “But we’re questioning from a legal perspective at what point does that need to have legal counsel review.”

McAndrew said it’s probably time for experts from various corners of the pen testing community to collaborate in documenting best practices that might help others avoid a repeat of the scenario in Dallas County.

“There’s no standard in the industry,” he said. “When it comes to these sorts of issues in red teaming — the legal challenges and the contracts — there’s really nothing out there. There are some things that can’t be undone. There’s the mugshots that are out there forever, but even as we get the charges dropped, these are permanently going to be in the federal database. This is a permanent thing that will reside with them and there’s no legal way we’re aware of to get these charges removed from the federal database.”

McAndrew said while he remains frustrated that it took so long to resolve this dispute, he doesn’t believe anyone involved acted with malicious intent.

“I don’t think there were any bad people,” he said. “Everyone was trying to do the right things — from law enforcement to the sheriff to the judges to the county — they all had the right intentions. But they didn’t necessarily all have the right information, and possibly people made decisions at levels they weren’t really authorized to do. Normally that’s not really our call, but I think people need to be thinking about that.”

Fresh on the heels of a disclosure that Microsoft Corp. leaked internal customer support data to the Internet, mobile provider Sprint has addressed a mix-up in which posts to a private customer support community were exposed to the Web.

KrebsOnSecurity recently contacted Sprint to let the company know that an internal customer support forum called “Social Care” was being indexed by search engines, and that several months worth of postings about customer complaints and other issues were viewable without authentication to anyone with a Web browser.

A Sprint spokesperson responded that the forum was indeed intended to be a private section of its support community, but that an error caused the section to become public.

“These conversations include minimal customer information and are used for frontline reps to escalate issues to managers,” said Lisa Belot, Sprint’s communications manager.

A review of the exposed support forum by this author suggests that while none of the posts exposed customer information such as payment card data, a number of them did include customer account information, such customer names, device identifiers and in some cases location information.

Perhaps more importantly for Sprint and its customers, the forum also included numerous links and references to internal tools and procedures. This sort of information would no doubt be of interest to scammers seeking to conduct social engineering attacks against Sprint employees as way to perpetrate other types of fraud, including unauthorized SIM swaps or in gleaning more account information from targeted customers.

Earlier this week, vice.com reported that hackers are phishing workers at major U.S. telecommunications companies to gain access to internal company tools. That news followed a related Vice report earlier this month which found ne’er-do-wells are now getting telecom employees to run software that lets the hackers directly reach into the internal systems of U.S. telecom companies to take over customer cell phone numbers.

The misstep by Sprint comes just days after Microsoft acknowledged that a database containing “a subset of information related to customer support interactions was accessible to the internet between the dates of Dec. 5 and Dec. 31, 2019.” Microsoft said it was alerting individuals whose information was exposed, which included location information, email and IP addresses, telephone numbers and descriptions of technical issues.

This week marked the annual observance of Data Privacy Day, an occasion in which we are reminded to be more judicious about the types of personal information we voluntarily share on social media and other Web sites. But both the Microsoft and Sprint stumbles are a reminder that billion-dollar companies very often expose this information on our behalf, even when we are doing everything within our power to safeguard it.