CISA Order Highlights Persistent Risk at Network Edge

The U.S. government agency in charge of improving the nation’s cybersecurity posture is ordering all federal agencies to take new measures to restrict access to Internet-exposed networking equipment. The directive comes amid a surge in attacks targeting previously unknown vulnerabilities in widely used security and networking appliances.

Under a new order from the Cybersecurity and Infrastructure Security Agency (CISA), federal agencies will have 14 days to respond to any reports from CISA about misconfigured or Internet-exposed networking equipment. The directive applies to any networking devices — such as firewalls, routers and load balancers — that allow remote authentication or administration.

The order requires federal departments to limit access so that only authorized users on an agency’s local or internal network can reach the management interfaces of these devices. CISA’s mandate follows a slew of recent incidents wherein attackers exploited zero-day flaws in popular networking products to conduct ransomware and cyber espionage attacks on victim organizations.

Earlier today, incident response firm Mandiant revealed that since at least October 2022, Chinese cyber spies have been exploiting a zero-day vulnerability in many email security gateway (ESG) appliances sold by California-based Barracuda Networks to hoover up email from organizations using these devices.

Barracuda was alerted to the exploitation of a zero-day in its products in mid-May, and two days later the company pushed a security update to address the flaw in all affected devices. But last week, Barracuda took the highly unusual step of offering to replace compromised ESGs, evidently in response to malware that altered the systems in such a fundamental way that they could no longer be secured remotely with software updates.

According to Mandiant, a previously unidentified Chinese hacking group was responsible for exploiting the Barracuda flaw, and appeared to be searching through victim organization email records for accounts “belonging to individuals working for a government with political or strategic interest to [China] while this victim government was participating in high-level, diplomatic meetings with other countries.”

When security experts began raising the alarm about a possible zero-day in Barracuda’s products, the Chinese hacking group altered their tactics, techniques and procedures (TTPs) in response to Barracuda’s efforts to contain and remediate the incident, Mandiant found.

Mandiant said the attackers will continue to change their tactics and malware, “especially as network defenders continue to take action against this adversary and their activity is further exposed by the infosec community.”

Meanwhile, this week we learned more details about the ongoing exploitation of a zero-day flaw in a broad range of virtual private networking (VPN) products made by Fortinet — devices many organizations rely on to facilitate remote network access for employees.

On June 11, Fortinet released a half-dozen security updates for its FortiOS firmware, including a weakness that researchers said allows an attacker to run malware on virtually any Fortinet SSL VPN appliance. The researchers found that just being able to reach the management interface for a vulnerable Fortinet SSL VPN appliance was enough to completely compromise the devices.

“This is reachable pre-authentication, on every SSL VPN appliance,” French vulnerability researcher Charles Fol tweeted. “Patch your #Fortigate.”

In details published on June 12, Fortinet confirmed that one of the vulnerabilities (CVE-2023-27997) is being actively exploited. The company said it discovered the weakness in an internal code audit that began in January 2023 — when it learned that Chinese hackers were exploiting a different zero-day flaw in its products.

Shodan.io, the search engine made for finding Internet of Things devices, reports that there are currently more than a half-million vulnerable Fortinet devices reachable via the public Internet.

The new cybersecurity directive from CISA orders agencies to remove any networking device management interfaces from the internet by making them only accessible from an internal enterprise network (CISA recommends an isolated management network). CISA also says agencies should “deploy capabilities, as part of a Zero Trust Architecture, that enforce access control to the interface through a policy enforcement point separate from the interface itself (preferred action).”

Security experts say CISA’s directive highlights the reality that cyberspies and ransomware gangs are making it increasingly risky for organizations to expose any devices to the public Internet, because these groups have strong incentives to probe such devices for previously unknown security vulnerabilities.

The most glaring example of this dynamic can be seen in the frequency with which ransomware groups have discovered and pounced on zero-day flaws in widely-used file transfer applications. One ransomware gang in particular — Cl0p — has repeatedly exploited zero day bugs in various file transfer appliances to extort tens of millions of dollars from hundreds of ransomware victims.

On February 2, KrebsOnSecurity broke the news that attackers were exploiting a zero-day vulnerability in the GoAnywhere file transfer appliance by Fortra. By the time security updates were available to fix the vulnerability, Cl0p had already used it to steal data from more than a hundred organizations running Fortra’s appliance.

According to CISA, on May 27, Cl0p began exploiting a previously unknown flaw in MOVEit Transfer, a popular Internet-facing file transfer application. MOVEit parent Progress Software has since released security updates to address the weakness, but Cl0p claims to have already used it to compromise hundreds of victim organizations. TechCrunch has been tracking the fallout from victim organizations, which range from banks and insurance providers to universities and healthcare entities.

The always on-point weekly security news podcast Risky Business has recently been urging organizations to jettison any and all FTP appliances, noting that Cl0p (or another crime gang) is likely to visit the same treatment on other FTP appliance vendors.

But that sound advice doesn’t exactly scale for mid-tier networking devices like Barracuda ESGs or Fortinet SSL VPNs, which are particularly prominent in small to mid-sized organizations.

“It’s not like FTP services, you can’t tell an enterprise [to] turn off the VPN [because] the productivity hit of disconnecting the VPN is terminal, it’s a non-starter,” Risky Business co-host Adam Boileau said on this week’s show. “So how to mitigate the impact of having to use a domain-joined network appliance at the edge of your network that is going to get zero-day in it? There’s no good answer.”

Risky Business founder Patrick Gray said the COVID-19 pandemic breathed new life into entire classes of networking appliances that rely on code which was never designed with today’s threat models in mind.

“In the years leading up to the pandemic, the push towards identity-aware proxies and zero trust everything and moving away from this type of equipment was gradual, but it was happening,” Gray said. “And then COVID-19 hit and everybody had to go work from home, and there really was one option to get going quickly — which was to deploy VPN concentrators with enterprise features.”

Gray said the security industry had been focused on building the next generation of remote access tools that are more security-hardened, but when the pandemic hit organizations scrambled to cobble together whatever they could.

“The only stuff available in the market was all this old crap that is not QA’d properly, and every time you shake them CVEs fall out,” Gray remarked, calling the pandemic, “a shot in the arm” to companies like Fortinet and Barracuda.

“They sold so many VPNs through the pandemic and this is the hangover,” Gray said. “COVID-19 extended the life of these companies and technologies, and that’s unfortunate.”

Service Rents Email Addresses for Account Signups

One of the most expensive aspects of any cybercriminal operation is the time and effort it takes to constantly create large numbers of new throwaway email accounts. Now a new service offers to help dramatically cut costs associated with large-scale spam and account creation campaigns, by paying people to sell their email account credentials and letting customers temporarily rent access to a vast pool of established accounts at major providers.

The service in question — kopeechka[.]store — is perhaps best described as a kind of unidirectional email confirmation-as-a-service that promises to “save your time and money for successfully registering multiple accounts.”

“Are you working on large volumes and are costs constantly growing?” Kopeechka’s website asks. “Our service will solve all your problems.”

As a customer of this service, you don’t get full access to the email inboxes you are renting. Rather, you configure your botnet or spam machine to make an automated application programming interface (API) call to the Kopeechka service, which responds with a working email address at an email provider of your choosing.

Once you’ve entered the supplied email address into the new account registration page at some website or service, you tell Kopeechka which service or website you’re expecting an account confirmation link from, and they will then forward any new messages matching that description to your Kopeechka account panel.

Ensuring that customers cannot control inboxes rented through the service means that Kopeechka can rent the same email address to multiple customers (at least until that email address has been used to register accounts at most of the major online services).

Kopeechka also has multiple affiliate programs, including one that pays app developers for embedding Kopeechka’s API in their software. However, far more interesting is their program for rewarding people who choose to sell Kopeechka usernames and passwords for working email addresses.

Kopeechka means “penny” in Russian, which is generous verbiage (and coinage) for a service that charges a tiny fraction of a penny for access to account confirmation links. Their pricing fluctuates slightly based on which email provider you choose, but a form on the service’s homepage says a single confirmation message from apple.com to outlook.com costs .07 rubles, which is currently equal to about $0.00087 dollars.

“Emails can be uploaded to us for sale, and you will receive a percentage of purchases %,” the service explains. “You upload 1 mailbox of a certain domain, discuss percentage with our technical support (it depends on the liquidity of the domain and the number of downloaded emails).”

We don’t have to look very far for examples of Kopeechka in action. In May, KrebsOnSecurity interviewed a Russian spammer named “Quotpw“ who was mass-registering accounts on the social media network Mastodon in order to conduct a series of huge spam campaigns advertising scam cryptocurrency investment platforms.

Much of the fodder for that story came from Renaud Chaput, a freelance programmer working on modernizing and scaling the Mastodon project infrastructure — including joinmastodon.org, mastodon.online, and mastodon.social. Chaput told KrebsOnSecurity that his team was forced to temporarily halt all new registrations for these communities last month after the number of new registrations from Quotpw’s spam campaign started to overwhelm their systems.

“We suddenly went from like three registrations per minute to 900 a minute,” Chaput said. “There was nothing in the Mastodon software to detect that activity, and the protocol is not designed to handle this.”

After that story ran, Chaput said he discovered that the computer code powering Quotpw’s spam botnet (which has since been released as open source) contained an API call to Kopeechka’s service.

“It allows them to pool many bot-created or compromised emails at various providers and offer them to cyber criminals,” Chaput said of Kopeechka. “This is what they used to create thousands of valid Hotmail (and other) addresses when spamming on Mastodon. If you look at the code, it’s really well done with a nice API that forwards you the confirmation link that you can then fake click with your botnet.”

It’s doubtful anyone will make serious money selling email accounts to Kopeechka, unless of course that person already happens to run a botnet and has access to ridiculous numbers of email credentials. And in that sense, this service is genius: It essentially offers scammers a new way to wring extra income from resources that are already plentiful for them.

One final note about Quotpw and the spam botnet that ravaged Chaput’s Mastodon servers last month: Trend Micro just published a report saying Quotpw was spamming to earn money for a Russian-language affiliate program called “Impulse Team,” which pays people to promote cryptocurrency scams.

Websites under the banner of the Impulse Scam Crypto Project are all essentially “advanced fee” scams that tell people they have earned a cryptocurrency investment credit. Upon registering at the site, visitors are told they need to make a minimum deposit on the service to collect the award. However, those who make the initial investment never hear from the site again, and their money is gone.

Interestingly, Trend Micro says the scammers behind the Impulse Team also appear to be operating a fake reputation service called Scam-Doc[.]com, a website that mimics the legitimate Scamdoc.com for measuring the trustworthiness and authenticity of various sites. Trend notes that the phony reputation site routinely gave high trust ratings to a variety of cryptocurrency scam and casino websites.

“We can only suppose that either the same cybercriminals run operations involving both or that several different cybercriminals share the scam-doc[.]com site,” the Trend researchers wrote.

According to the FBI, financial losses from cryptocurrency investment scams dwarfed losses for all other types of cybercrime in 2022, rising from $907 million in 2021 to $2.57 billion last year.

Discord Admins Hacked by Malicious Bookmarks

A number of Discord communities focused on cryptocurrency have been hacked this past month after their administrators were tricked into running malicious Javascript code disguised as a Web browser bookmark.

According to interviews with victims, several of the attacks began with an interview request from someone posing as a reporter for a crypto-focused news outlet online. Those who take the bait are sent a link to a Discord server that appears to be the official Discord of the crypto news site, where they are asked to complete a verification step to validate their identity.

As shown in this Youtube video, the verification process involves dragging a button from the phony crypto news Discord server to the bookmarks bar in one’s Web browser. From there, the visitor is instructed to go back to discord.com and then click the new bookmark to complete the verification process.

However, the bookmark is actually a clever snippet of Javascript that quietly grabs the user’s Discord token and sends it to the scammer’s website. The attacker then loads the stolen token into their own browser session and (usually late at night after the admins are asleep) posts an announcement in the targeted Discord about an exclusive “airdrop,” “NFT mint event” or some other potential money making opportunity for the Discord members.

The unsuspecting Discord members click the link provided by the compromised administrator account, and are asked to connect their crypto wallet to the scammer’s site, where it asks for unlimited spend approvals on their tokens, and subsequently drains the balance of any valuable accounts.

Meanwhile, anyone in the compromised Discord channel who notices the scam and replies is banned, and their messages are deleted by the compromised admin account.

Nicholas Scavuzzo is an associate at Ocean Protocol, which describes itself as an “open-source protocol that aims to allow businesses and individuals to exchange and monetize data and data-based services.” On May 22, an administrator for Ocean Protocol’s Discord server clicked a link in a direct message from a community member that prompted them to prove their identity by dragging a link to their bookmarks.

Scavuzzo, who is based in Maine, said the attackers waited until around midnight in his timezone time before using the administrator’s account to send out an unauthorized message about a new Ocean airdrop.

Scavuzzo said the administrator’s account was hijacked even though she had multi-factor authentication turned on.

“A CAPTCHA bot that allows Discord cookies to be accessed by the person hosting the CAPTCHA,” was how Scavuzzo described the attack. “I’ve seen all kinds of crypto scams, but I’ve never seen one like this.”

Importantly, the stolen token only works for the attackers as long as its rightful owner doesn’t log out and back in, or else change their credentials.

Assuming the administrator can log in, that is. In Ocean’s case, one of the first things the intruders did once they swiped the administrator’s token was change the server’s access controls and remove all core Ocean team members from the server.

Fortunately for Ocean, Scavuzzo was able to reach the operator of the server that hosts the Discord channel, and have the channel’s settings reverted back to normal.

“Thankfully, we are a globally distributed team, so we have people awake at all hours,” Scavuzzo said, noting that Ocean is not aware of any Discord community members who fell for the phony airdrop offer, which was live for about 30 minutes. “This could have been a lot worse.”

On May 26, Aura Network reported on Twitter that its Discord server was compromised in a phishing attack that resulted in the deletion of Discord channels and the dissemination of fake Aura Network Airdrop Campaign links.

On May 27, Nahmii — a cryptocurrency technology based on the Ethereum blockchain — warned on Twitter that one of its community moderators on Discord was compromised and posting fake airdrop details.

On May 9, MetrixCoin reported that its Discord server was hacked, with fake airdrop details pushed to all users.

KrebsOnSecurity recently heard from a trusted source in the cybersecurity industry who dealt firsthand with one of these attacks and asked to remain anonymous.

“I do pro bono Discord security work for a few Discords, and I was approached by one of these fake journalists,” the source said. “I played along and got the link to their Discord, where they were pretending to be journalists from the Cryptonews website using several accounts.”

The source took note of all the Discord IDs of the admins of the fake Cryptonews Discord, so that he could ensure they were blocked from the Discords he helps to secure.

“Since I’ve been doing this for a while now, I’ve built up a substantial database of Discord users and messages, so often I can see these scammers’ history on Discord,” the source said.

In this case, he noticed a user with the “CEO” role in the fake Cryptonews Discord had been seen previously under another username — “Levatax.” Searching on that Discord ID and username revealed a young Turkish coder named Berk Yilmaz whose Github page linked to the very same Discord ID as the scammer CEO.

Reached via instant message on Telegram, Levatax said he’s had no involvement in such schemes, and that he hasn’t been on Discord since his Microsoft Outlook account was hacked months ago.

“The interesting thing [is] that I didn’t use Discord since few months or even social media because of the political status of Turkey,” Levatax explained, referring to the recent election in his country. “The only thing I confirm is losing my Outlook account which connected to my Discord, and I’m already in touch with Microsoft to recover it.”

The verification method used in the above scam involves a type of bookmark called a “bookmarklet” that stores Javascript code as a clickable link in the bookmarks bar at the top of one’s browser.

While bookmarklets can be useful and harmless, malicious Javascript that is executed in the browser by the user is especially dangerous. So please avoid adding (or dragging) any bookmarks or bookmarklets to your browser unless it was your idea in the first place.

Interview With a Crypto Scam Investment Spammer

Social networks are constantly battling inauthentic bot accounts that send direct messages to users promoting scam cryptocurrency investment platforms. What follows is an interview with a Russian hacker responsible for a series of aggressive crypto spam campaigns that recently prompted several large Mastodon communities to temporarily halt new registrations. According to the hacker, their spam software has been in private use until the last few weeks, when it was released as open source code.

Renaud Chaput is a freelance programmer working on modernizing and scaling the Mastodon project infrastructure — including joinmastodon.org, mastodon.online, and mastodon.social. Chaput said that on May 4, 2023, someone unleashed a spam torrent targeting users on these Mastodon communities via “private mentions,” a kind of direct messaging on the platform.

The messages said recipients had earned an investment credit at a cryptocurrency trading platform called moonxtrade[.]com. Chaput said the spammers used more than 1,500 Internet addresses across 400 providers to register new accounts, which then followed popular accounts on Mastodon and sent private mentions to the followers of those accounts.

Since then, the same spammers have used this method to advertise more than 100 different crypto investment-themed domains. Chaput said that at one point this month the volume of bot accounts being registered for the crypto spam campaign started overwhelming the servers that handle new signups at Mastodon.social.

“We suddenly went from like three registrations per minute to 900 a minute,” Chaput said. “There was nothing in the Mastodon software to detect that activity, and the protocol is not designed to handle this.”

Seeking to gain a temporary handle on the spam wave, Chaput said he briefly disabled new account registrations on mastodon.social and mastondon.online. Shortly after that, those same servers came under a sustained distributed denial-of-service (DDoS) attack.

Chaput said whoever was behind the DDoS was definitely not using point-and-click DDoS tools, like a booter or stresser service.

“This was three hours non-stop, 200,000 to 400,000 requests per second,” Chaput said of the DDoS. “At first, they were targeting one path, and when we blocked that they started to randomize things. Over three hours the attack evolved several times.”

Chaput says the spam waves have died down since they retrofitted mastodon.social with a CAPTCHA, those squiggly letter and number combinations designed to stymie automated account creation tools. But he’s worried that other Mastodon instances may not be as well-staffed and might be easy prey for these spammers.

“We don’t know if this is the work of one person, or if this is [related to] software or services being sold to others,” Chaput told KrebsOnSecurity. “We’re really impressed by the scale of it — using hundreds of domains and thousands of Microsoft email addresses.”

Chaput said a review of their logs indicates many of the newly registered Mastodon spam accounts were registered using the same 0auth credentials, and that a domain common to those credentials was quot[.]pw.

A DIRECT QUOT

The domain quot[.]pw has been registered and abandoned by several parties since 2014, but the most recent registration data available through DomainTools.com shows it was registered in March 2020 to someone in Krasnodar, Russia with the email address edgard011012@gmail.com.

This email address is also connected to accounts on several Russian cybercrime forums, including “__edman__,” who had a history of selling “logs” — large amounts of data stolen from many bot-infected computers — as well as giving away access to hacked Internet of Things (IoT) devices.

In September 2018, a user by the name “ципа” (phonetically “Zipper” in Russian) registered on the Russian hacking forum Lolzteam using the edgard0111012@gmail.com address. In May 2020, Zipper told another Lolzteam member that quot[.]pw was their domain. That user advertised a service called “Quot Project” which said they could be hired to write programming scripts in Python and C++.

“I make Telegram bots and other rubbish cheaply,” reads one February 2020 sales thread from Zipper.

Clicking the “open chat in Telegram” button on Zipper’s Lolzteam profile page launched a Telegram instant message chat window where the user Quotpw responded almost immediately. Asked if they were aware their domain was being used to manage a spam botnet that was pelting Mastodon instances with crypto scam spam, Quotpw confirmed the spam was powered by their software.

“It was made for a limited circle of people,” Quotpw said, noting that they recently released the bot software as open source on GitHub.

Quotpw went on to say the spam botnet was powered by well more than the hundreds of IP addresses tracked by Chaput, and that these systems were mostly residential proxies. A residential proxy generally refers to a computer or mobile device running some type of software that enables the system to be used as a pass-through for Internet traffic from others.

Very often, this proxy software is installed surreptitiously, such as through a “Free VPN” service or mobile app. Residential proxies also can refer to households protected by compromised home routers running factory-default credentials or outdated firmware.

Quotpw maintains they have earned more than $2,000 sending roughly 100,000 private mentions to users of different Mastodon communities over the past few weeks. Quotpw said their conversion rate for the same bot-powered direct message spam on Twitter is usually much higher and more profitable, although they conceded that recent adjustments to Twitter’s anti-bot CAPTCHA have put a crimp in their Twitter earnings.

“My partners (I’m programmer) lost time and money while ArkoseLabs (funcaptcha) introduced new precautions on Twitter,” Quotpw wrote in a Telegram reply. “On Twitter, more spam and crypto scam.”

Asked whether they felt at all conflicted about spamming people with invitations to cryptocurrency scams, Quotpw said in their hometown “they pay more for such work than in ‘white’ jobs” — referring to legitimate programming jobs that don’t involve malware, botnets, spams and scams.

“Consider salaries in Russia,” Quotpw said. “Any spam is made for profit and brings illegal money to spammers.”

THE VIENNA CONNECTION

Shortly after edgard011012@gmail.com registered quot[.]pw, the WHOIS registration records for the domain were changed again, to msr-sergey2015@yandex.ru, and to a phone number in Austria: +43.6607003748.

Constella Intelligence, a company that tracks breached data, finds that the address msr-sergey2015@yandex.ru has been associated with accounts at the mobile app site aptoide.com (user: CoolappsforAndroid) and vimeworld.ru that were created from different Internet addresses in Vienna, Austria.

A search in Skype on that Austrian phone number shows it belongs to a Sergey Proshutinskiy who lists his location as Vienna, Austria. The very first result that comes up when one searches that unusual name in Google is a LinkedIn profile for a Sergey Proshutinskiy from Vienna, Austria.

Proshutinskiy’s LinkedIn profile says he is a Class of 2024 student at TGM, which is a state-owned, technical and engineering school in Austria. His resume also says he is a data science intern at Mondi Group, an Austrian manufacturer of sustainable packaging and paper.

Mr. Proshutinskiy did not respond to requests for comment.

Quotpw denied being Sergey, and said Sergey was a friend who registered the domain as a birthday present and favor last year.

“Initially, I bought it for 300 rubles,” Quotpw explained. “The extension cost 1300 rubles (expensive). I waited until it expired and forgot to buy it. After that, a friend (Sergey) bought [the] domain and transferred access rights to me.”

“He’s not even an information security specialist,” Quotpw said of Sergey. “My friends do not belong to this field. None of my friends are engaged in scams or other black [hat] activities.”

It may seem unlikely that someone would go to all this trouble to spam Mastodon users over several weeks using an impressive number of resources — all for just $2,000 in profit. But it is likely that whoever is actually running the various crypto scam platforms advertised by Quotpw’s spam messages pays handsomely for any investments generated by their spam.

According to the FBI, financial losses from cryptocurrency investment scams dwarfed losses for all other types of cybercrime in 2022, rising from $907 million in 2021 to $2.57 billion last year.

Update, May 25, 10:30 a.m.:  Corrected attribution of the Austrian school TGM.

Re-Victimization from Police-Auctioned Cell Phones

Countless smartphones seized in arrests and searches by police forces across the United States are being auctioned online without first having the data on them erased, a practice that can lead to crime victims being re-victimized, a new study found. In response, the largest online marketplace for items seized in U.S. law enforcement investigations says it now ensures that all phones sold through its platform will be data-wiped prior to auction.

Researchers at the University of Maryland last year purchased 228 smartphones sold “as-is” from PropertyRoom.com, which bills itself as the largest auction house for police departments in the United States. Of phones they won at auction (at an average of $18 per phone), the researchers found 49 had no PIN or passcode; they were able to guess an additional 11 of the PINs by using the top-40 most popular PIN or swipe patterns.

Phones may end up in police custody for any number of reasons — such as its owner was involved in identity theft — and in these cases the phone itself was used as a tool to commit the crime.

“We initially expected that police would never auction these phones, as they would enable the buyer to recommit the same crimes as the previous owner,” the researchers explained in a paper released this month. “Unfortunately, that expectation has proven false in practice.”

The researchers said while they could have employed more aggressive technological measures to work out more of the PINs for the remaining phones they bought, they concluded based on the sample that a great many of the devices they won at auction had probably not been data-wiped and were protected only by a PIN.

Beyond what you would expect from unwiped second hand phones — every text message, picture, email, browser history, location history, etc. — the 61 phones they were able to access also contained significant amounts of data pertaining to crime — including victims’ data — the researchers found.

Some readers may be wondering at this point, “Why should we care about what happens to a criminal’s phone?” First off, it’s not entirely clear how these phones ended up for sale on PropertyRoom.

“Some folks are like, ‘Yeah, whatever, these are criminal phones,’ but are they?” said Dave Levin, an assistant professor of computer science at University of Maryland.

“We started looking at state laws around what they’re supposed to do with lost or stolen property, and we found that most of it ends up going the same route as civil asset forfeiture,” Levin continued. “Meaning, if they can’t find out who owns something, it eventually becomes the property of the state and gets shipped out to these resellers.”

Also, the researchers found that many of the phones clearly had personal information on them regarding previous or intended targets of crime: A dozen of the phones had photographs of government-issued IDs. Three of those were on phones that apparently belonged to sex workers; their phones contained communications with clients.

One phone had full credit files for eight different people on it. On another device they found a screenshot including 11 stolen credit cards that were apparently purchased from an online carding shop. On yet another, the former owner had apparently been active in a Telegram group chat that sold tutorials on how to run identity theft scams.

The most interesting phone from the batches they bought at auction was one with a sticky note attached that included the device’s PIN and the notation “Gry Keyed,” no doubt a reference to the Graykey software that is often used by law enforcement agencies to brute-force a mobile device PIN.

“That one had the PIN on the back,” Levin said. “The message chain on that phone had 24 Experian and TransUnion credit histories”.

The University of Maryland team said they took care in their research not to further the victimization of people whose information was on the devices they purchased from PropertyRoom.com. That involved ensuring that none of the devices could connect to the Internet when powered on, and scanning all images on the devices against known hashes for child sexual abuse material.

It is common to find phones and other electronics for sale on auction platforms like eBay that have not been wiped of sensitive data, but in those cases eBay doesn’t possess the items being sold. In contrast, platforms like PropertyRoom obtain devices and resell them at auction directly.

PropertyRoom did not respond to multiple requests for comment. But the researchers said sometime in the past few months PropertyRoom began posting a notice stating that all mobile devices would be wiped of their data before being sold at auction.

“We informed them of our research in October 2022, and they responded that they would review our findings internally,” Levin said. “They stopped selling them for a while, but then it slowly came back, and then we made sure we won every auction. And all of the ones we got from that were indeed wiped, except there were four devices that had external SD [storage] cards in them that weren’t wiped.”

A copy of the University of Maryland study is here (PDF).

Giving a Face to the Malware Proxy Service ‘Faceless’

For the past seven years, a malware-based proxy service known as “Faceless” has sold anonymity to countless cybercriminals. For less than a dollar per day, Faceless customers can route their malicious traffic through tens of thousands of compromised systems advertised on the service. In this post we’ll examine clues left behind over the past decade by the proprietor of Faceless, including some that may help put a face to the name.

Riley Kilmer is co-founder of Spur.us, a company that tracks thousands of VPN and proxy networks, and helps customers identify traffic coming through these anonymity services. Kilmer said Faceless has emerged as one of the underground’s most reliable malware-based proxy services, mainly because its proxy network has traditionally included a great many compromised “Internet of Things” devices — such as media sharing servers — that are seldom included on malware or spam block lists.

Kilmer said when Spur first started looking into Faceless, they noticed almost every Internet address that Faceless advertised for rent also showed up in the IoT search engine Shodan.io as a media sharing device on a local network that was somehow exposed to the Internet.

“We could reliably look up the [fingerprint] for these media sharing devices in Shodan and find those same systems for sale on Faceless,” Kilmer said.

In January 2023, the Faceless service website said it was willing to pay for information about previously undocumented security vulnerabilities in IoT devices. Those with IoT zero-days could expect payment if their exploit involved at least 5,000 systems that could be identified through Shodan.

Recently, Faceless has shown ambitions beyond just selling access to poorly-secured IoT devices. In February, Faceless re-launched a service that lets users drop an email bomb on someone — causing the target’s inbox to be filled with tens of thousands of junk messages.

And in March 2023, Faceless started marketing a service for looking up Social Security Numbers (SSNs) that claims to provide access to “the largest SSN database on the market with a very high hit rate.”

Kilmer said Faceless wants to become a one-stop-fraud-shop for cybercriminals who are seeking stolen or synthetic identities from which to transact online, and a temporary proxy that is geographically close to the identity being sold. Faceless currently sells this bundled product for $9 — $8 for the identity and $1 for the proxy.

“They’re trying to be this one-stop shop for anonymity and personas,” Kilmer said. “The service basically says ‘here’s an SSN and proxy connection that should correspond to that user’s location and make sense to different websites.'”

MRMURZA

Faceless is a project from MrMurza, a particularly talkative member of more than a dozen Russian-language cybercrime forums over the past decade. According to cyber intelligence firm Flashpoint, MrMurza has been active in the Russian underground since at least September 2012. Flashpoint said MrMurza appears to be extensively involved in botnet activity and “drops” — fraudulent bank accounts created using stolen identity data that are often used in money laundering and cash-out schemes.

Faceless grew out of a popular anonymity service called iSocks, which was launched in 2014 and advertised on multiple Russian crime forums as a proxy service that customers could use to route their malicious Web traffic through compromised computers.

Flashpoint says that in the months before iSocks went online, MrMurza posted on the Russian language crime forum Verified asking for a serious partner to assist in opening a proxy service, noting they had a botnet that was powered by malware that collected proxies with a 70 percent infection rate.

In September 2016, MrMurza sent a message to all iSocks users saying the service would soon be phased out in favor of Faceless, and that existing iSocks users could register at Faceless for free if they did so quickly — before Faceless began charging new users registration fees between $50 and $100.

Verified and other Russian language crime forums where MrMurza had a presence have been hacked over the years, with contact details and private messages leaked online. In a 2014 private message to the administrator of Verified explaining his bona fides, MrMurza said he received years of positive feedback as a seller of stolen Italian credit cards and a vendor of drops services.

MrMurza told the Verified admin that he used the nickname AccessApproved on multiple other forums over the years. MrMurza also told the admin that his account number at the now-defunct virtual currency Liberty Reserve was U1018928.

According to cyber intelligence firm Intel 471, the user AccessApproved joined the Russian crime forum Zloy in Jan. 2012, from an Internet address in Magnitogorsk, RU. In a 2012 private message where AccessApproved was arguing with another cybercriminal over a deal gone bad, AccessApproved asked to be paid at the Liberty Reserve address U1018928.

In 2013, U.S. federal investigators seized Liberty Reserve and charged its founders with facilitating billions of dollars in money laundering tied to cybercrime. The Liberty Reserve case was prosecuted out of the Southern District of New York, which in 2016 published a list of account information (PDF) tied to thousands of Liberty Reserve addresses the government asserts were involved in money laundering.

That document indicates the Liberty Reserve account claimed by MrMurza/AccessApproved — U1018928 — was assigned in 2011 to a “Vadim Panov” who used the email address lesstroy@mgn.ru.

PANOV

Constella Intelligence, a threat intelligence firm that tracks breached databases, says lesstroy@mgn.ru was used for an account “Hackerok” at the accounting service klerk.ru that was created from an Internet address in Magnitogorsk. The password chosen by this user was “1232.”

In addition to selling access to hacked computers and bank accounts, both MrMurza and AccessApproved ran side hustles on the crime forums selling clothing from popular retailers that refused to ship directly to Russia.

On one cybercrime forum where AccessApproved had clothing customers, denizens of the forum created a lengthy discussion thread to help users identify incoming emails associated with various reshipping services advertised within their community. Reshippers tend to rely on a large number of people in the United States and Europe helping to forward packages overseas, but in many cases the notifications about purchases and shipping details would be forwarded to reshipping service customers from a consistent email account.

That thread said AccessApproved’s clothing reshipping service forwarded confirmation emails from the address panov-v@mail.ru. This address is associated with accounts on two Russian cybercrime forums registered from Magnitogorsk in 2010 using the handle “Omega^gg4u.”

This Omega^gg4u identity sold software that can rapidly check the validity of large batches of stolen credit cards. Interestingly, both Omega^gg4u and AccessApproved also had another niche: Reselling heavily controlled substances — such as human growth hormone and anabolic steroids — from chemical suppliers in China.

A search in Constella on the address panov-v@mail.ru and many variations on that address shows these accounts cycled through the same passwords, including 055752403k, asus666, 01091987h, and the relatively weak password 1232 (recall that 1232 was picked by whoever registered the lesstroy@mgn.ru account at Klerk.ru).

Constella says the email address asus666@yandex.ru relied on the passwords asus666 and 01091987h. The 01091987h password also was used by asus666@mail.ru, which also favored the password 24587256.

Constella further reports that whoever owned the much shorter address asus@mail.ru also used the password 24587256. In addition, it found the password 2318922479 was tied to both asus666@mail.ru and asus@mail.ru.

The email addresses asus@mail.ru, asus2504@mail.ru, and zaxar2504@rambler.ru were all used to register Vkontakte social media accounts for a Denis ***@VIP*** Pankov. There are a number of other Vkontakte accounts registered to asus@mail.ru and many variations of this address under a different name. But none of those other profiles appear tied to real-life identities.

PANKOV

Constella’s data shows the email addresses asus2504@mail.ru and zaxar2504@rambler.ru used the rather unique password denis250485, which was also used by the email address denispankov@yandex.ru and almost a dozen variations at other Russian-language email providers.

Russian vehicle registration records from 2016 show the email address denispankov@yandex.ru belongs to Denis Viktorovich Pankov, born on April 25, 1985. That explains the “250485” portion of Pankov’s favored password. The registration records further indicate that in 2016 Pankov’s vehicle was registered in a suburb of Moscow.

Russian incorporation records show that denispankov@yandex.com is tied to IP Pankov Denis Viktorovich, a now-defunct transportation company in the Volograd Oblast, a region in southern Russia that shares a long border with western Kazazkhstan.

More recent records for IP Pankov Denis Viktorovich show a microenterprise with this name in Omsk that described its main activity as “retail sale by mail or via the Internet.” Russian corporate records indicate this entity was liquidated in 2021.

A reverse password search on “denis250485” via Constella shows this password was used by more than 75 email addresses, most of which are some variation of gaihnik@mail.ru — such as gaihnik25@mail.ru, or gaihnik2504@rambler.ru.

In 2012, someone posted answers to a questionnaire on behalf of Denis Viktorovich Pankov to a Russian-language discussion forum on Chinese crested dog breeds. The message said Pankov was seeking a puppy of a specific breed and was a resident of Krasnogorsk, a city that is adjacent to the northwestern boundary of Moscow.

The message said Pankov was a then 27-year-old manager in an advertising company, and could be reached at the email address gaihnik@mail.ru.

GAIHNIK

Constella Intelligence shows gaihnik@mail.ru registered at the now-defunct email marketing service Smart Responder from an address in Gagarin, which is about 115 miles west of Moscow.

Back in 2015, the user Gaihnik25 was banned from the online game World of Tanks for violating the game’s terms that prohibit “bot farming,” or the automated use of large numbers of player accounts to win some advantage that is usually related to cashing out game accounts or inventory.

For the past few years, someone using the nickname Gaihnik25 has been posting messages to the Russian-language hacking forum Gerki[.]pw, on discussion threads regarding software designed to “brute force” or mass-check online accounts for weak or compromised passwords.

A new member of the Russian hacking forum Nohide[.]Space using the handle Gaihnik has been commenting recently about proxy services, credential checking software, and the sale of hacked mailing lists. Gaihnik’s first post on the forum concerned private software for checking World of Tanks accounts.

The address gaihnik@mail.ru shows how so many email addresses tied to Pankov were also connected to apparently misleading identities on Vkontakte and elsewhere. Constella found this address was tied to a Vkontakte account for a Dmitriy Zakarov.

Microsoft’s Bing search engine says gaihnik@mail.ru belongs to 37-year-old Denis Pankov, yet clicking the Mail.ru profile for that user brings up a profile for a much older man by the name Gavril Zakarov. However, when you log in to a Mail.ru account and view that profile, it shows that most of the account’s profile photos are of a much younger man.

Many of those same photos show up in an online dating profile at dating.ru for the user Gaihnik, a.k.a “Denchik,” who says he is a 37-year-old Taurus from Gagarin who enjoys going for walks in nature, staying up late, and being on the Internet.

Mr. Pankov did not respond to multiple requests for comment sent to all of the email addresses mentioned in this story. However, some of those addresses produced detailed error responses; Mail.ru reported that the users panov-v@mail.ru, asus666@mail.ru, and asus2504@mail.ru were terminated, and that gaihnik25@mail.ru is now disabled.

Messages sent to many other email addresses connected via passwords to Pankov and using some variation of asus####@mail.ru also returned similar account termination messages.

Google Suspends Chinese E-Commerce App Pinduoduo Over Malware

Google says it has suspended the app for the Chinese e-commerce giant Pinduoduo after malware was found in versions of the software. The move comes just weeks after Chinese security researchers published an analysis suggesting the popular e-commerce app sought to seize total control over affected devices by exploiting multiple security vulnerabilities in a variety of Android-based smartphones.

In November 2022, researchers at Google’s Project Zero warned about active attacks on Samsung mobile phones which chained together three security vulnerabilities that Samsung patched in March 2021, and which would have allowed an app to add or read any files on the device.

Google said it believes the exploit chain for Samsung devices belonged to a “commercial surveillance vendor,” without elaborating further. The highly technical writeup also did not name the malicious app in question.

On Feb. 28, 2023, researchers at the Chinese security firm DarkNavy published a blog post purporting to show evidence that a major Chinese ecommerce company’s app was using this same three-exploit chain to read user data stored by other apps on the affected device, and to make its app nearly impossible to remove.

DarkNavy likewise did not name the app they said was responsible for the attacks. In fact, the researchers took care to redact the name of the app from multiple code screenshots published in their writeup. DarkNavy did not respond to requests for clarification.

“At present, a large number of end users have complained on multiple social platforms,” reads a translated version of the DarkNavy blog post. “The app has problems such as inexplicable installation, privacy leakage, and inability to uninstall.”

Update, March 27, 1:24 p.m. ET: Dan Goodin over at Ars Technica has an important update on this story that indicates the Pinduoduo code was exploiting a zero-day vulnerability in Android — not Samsung. From that piece:

“A preliminary analysis by Lookout found that at least two off-Play versions of Pinduoduo for Android exploited CVE-2023-20963, the tracking number for an Android vulnerability Google patched in updates that became available to end users two weeks ago. This privilege-escalation flaw, which was exploited prior to Google’s disclosure, allowed the app to perform operations with elevated privileges. The app used these privileges to download code from a developer-designated site and run it within a privileged environment.

“The malicious apps represent “a very sophisticated attack for an app-based malware,” Christoph Hebeisen, one of three Lookout researchers who analyzed the file, wrote in an email. “In recent years, exploits have not usually been seen in the context of mass-distributed apps. Given the extremely intrusive nature of such sophisticated app-based malware, this is an important threat mobile users need to protect against.”

On March 3, 2023, a denizen of the now-defunct cybercrime community BreachForums posted a thread which noted that a unique component of the malicious app code highlighted by DarkNavy also was found in the ecommerce application whose name was apparently redacted from the DarkNavy analysis: Pinduoduo.

On March 4, 2023, e-commerce expert Liu Huafang posted on the Chinese social media network Weibo that Pinduoduo’s app was using security vulnerabilities to gain market share by stealing user data from its competitors. That Weibo post has since been deleted.

On March 7, the newly created Github account Davinci1010 published a technical analysis claiming that until recently Pinduoduo’s source code included a “backdoor,” a hacking term used to describe code that allows an adversary to remotely and secretly connect to a compromised system at will.

That analysis includes links to archived versions of Pinduoduo’s app released before March 5 (version 6.50 and lower), which is when Davinci1010 says a new version of the app removed the malicious code.

Pinduoduo has not yet responded to requests for comment. Pinduoduo parent company PDD Holdings told Reuters Google has not shared details about why it suspended the app.

The company told CNN that it strongly rejects “the speculation and accusation that Pinduoduo app is malicious just from a generic and non-conclusive response from Google,” and said there were “several apps that have been suspended from Google Play at the same time.”

Pinduoduo is among China’s most popular e-commerce platforms, boasting approximately 900 million monthly active users.

Most of the news coverage of Google’s move against Pinduoduo emphasizes that the malware was found in versions of the Pinduoduo app available outside of Google’s app store — Google Play.

“Off-Play versions of this app that have been found to contain malware have been enforced on via Google Play Protect,” a Google spokesperson said in a statement to Reuters, adding that the Play version of the app has been suspended for security concerns.

However, Google Play is not available to consumers in China. As a result, the app will still be available via other mobile app stores catering to the Chinese market — including those operated by Huawei, Oppo, Tencent and VIVO.

Google said its ban did not affect the PDD Holdings app Temu, which is an online shopping platform in the United States. According to The Washington Post, four of the Apple App Store’s 10 most-downloaded free apps are owned by Chinese companies, including Temu and the social media network TikTok.

The Pinduoduo suspension comes as lawmakers in Congress this week are gearing up to grill the CEO of TikTok over national security concerns. TikTok, which is owned by Beijing-based ByteDance, said last month that it now has roughly 150 million monthly active users in the United States.

A new cybersecurity strategy released earlier this month by the Biden administration singled out China as the greatest cyber threat to the U.S. and Western interests. The strategy says China now presents the “broadest, most active, and most persistent threat to both government and private sector networks,” and says China is “the only country with both the intent to reshape the international order and, increasingly, the economic, diplomatic, military, and technological power to do so.”

Why You Should Opt Out of Sharing Data With Your Mobile Provider

A new breach involving data from nine million AT&T customers is a fresh reminder that your mobile provider likely collects and shares a great deal of information about where you go and what you do with your mobile device — unless and until you affirmatively opt out of this data collection. Here’s a primer on why you might want to do that, and how.

Telecommunications giant AT&T disclosed this month that a breach at a marketing vendor exposed certain account information for nine million customers. AT&T said the data exposed did not include sensitive information, such as credit card or Social Security numbers, or account passwords, but was limited to “Customer Proprietary Network Information” (CPNI), such as the number of lines on an account.

Certain questions may be coming to mind right now, like “What the heck is CPNI?” And, ‘If it’s so ‘customer proprietary,’ why is AT&T sharing it with marketers?” Also maybe, “What can I do about it?” Read on for answers to all three questions.

AT&T’s disclosure said the information exposed included customer first name, wireless account number, wireless phone number and email address. In addition, a small percentage of customer records also exposed the rate plan name, past due amounts, monthly payment amounts and minutes used.

CPNI refers to customer-specific “metadata” about the account and account usage, and may include:

-Called phone numbers
-Time of calls
-Length of calls
-Cost and billing of calls
-Service features
-Premium services, such as directory call assistance

According to a succinct CPNI explainer at TechTarget, CPNI is private and protected information that cannot be used for advertising or marketing directly.

“An individual’s CPNI can be shared with other telecommunications providers for network operating reasons,” wrote TechTarget’s Gavin Wright. “So, when the individual first signs up for phone service, this information is automatically shared by the phone provider to partner companies.”

Is your mobile Internet usage covered by CPNI laws? That’s less clear, as the CPNI rules were established before mobile phones and wireless Internet access were common. TechTarget’s CPNI primer explains:

“Under current U.S. law, cellphone use is only protected as CPNI when it is being used as a telephone. During this time, the company is acting as a telecommunications provider requiring CPNI rules. Internet use, websites visited, search history or apps used are not protected CPNI because the company is acting as an information services provider not subject to these laws.”

Hence, the carriers can share and sell this data because they’re not explicitly prohibited from doing so. All three major carriers say they take steps to anonymize the customer data they share, but researchers have shown it is not terribly difficult to de-anonymize supposedly anonymous web-browsing data.

“Your phone, and consequently your mobile provider, know a lot about you,” wrote Jack Morse for Mashable. “The places you go, apps you use, and the websites you visit potentially reveal all kinds of private information — e.g. religious beliefs, health conditions, travel plans, income level, and specific tastes in pornography. This should bother you.”

Happily, all of the U.S. carriers are required to offer customers ways to opt out of having data about how they use their devices shared with marketers. Here’s a look at some of the carrier-specific practices and opt-out options.

AT&T

AT&T’s policy says it shares device or “ad ID”, combined with demographics including age range, gender, and ZIP code information with third parties which explicitly include advertisers, programmers, and networks, social media networks, analytics firms, ad networks and other similar companies that are involved in creating and delivering advertisements.

AT&T said the data exposed on 9 million customers was several years old, and mostly related to device upgrade eligibility. This may sound like the data went to just one of its partners who experienced a breach, but in all likelihood it also went to hundreds of AT&T’s partners.

AT&T’s CPNI opt-out page says it shares CPNI data with several of its affiliates, including WarnerMedia, DirecTV and Cricket Wireless. Until recently, AT&T also shared CPNI data with Xandr, whose privacy policy in turn explains that it shares data with hundreds of other advertising firms. Microsoft bought Xandr from AT&T last year.

T-MOBILE

According to the Electronic Privacy Information Center (EPIC), T-Mobile seems to be the only company out of the big three to extend to all customers the rights conferred by the California Consumer Privacy Act (CCPA).

EPIC says T-Mobile customer data sold to third parties uses another unique identifier called mobile advertising IDs or “MAIDs.” T-Mobile claims that MAIDs don’t directly identify consumers, but under the CCPA MAIDs are considered “personal information” that can be connected to IP addresses, mobile apps installed or used with the device, any video or content viewing information, and device activity and attributes.

T-Mobile customers can opt out by logging into their account and navigating to the profile page, then to “Privacy and Notifications.” From there, toggle off the options for “Use my data for analytics and reporting” and “Use my data to make ads more relevant to me.”

VERIZON

Verizon’s privacy policy says it does not sell information that personally identities customers (e.g., name, telephone number or email address), but it does allow third-party advertising companies to collect information about activity on Verizon websites and in Verizon apps, through MAIDs, pixels, web beacons and social network plugins.

According to Wired.com’s tutorial, Verizon users can opt out by logging into their Verizon account through a web browser or the My Verizon mobile app. From there, select the Account tab, then click Account Settings and Privacy Settings on the web. For the mobile app, click the gear icon in the upper right corner and then Manage Privacy Settings.

On the privacy preferences page, web users can choose “Don’t use” under the Custom Experience section. On the My Verizon app, toggle any green sliders to the left.

EPIC notes that all three major carriers say resetting the consumer’s device ID and/or clearing cookies in the browser will similarly reset any opt-out preferences (i.e., the customer will need to opt out again), and that blocking cookies by default may also block the opt-out cookie from being set.

T-Mobile says its opt out is device-specific and/or browser-specific. “In most cases, your opt-out choice will apply only to the specific device or browser on which it was made. You may need to separately opt out from your other devices and browsers.”

Both AT&T and Verizon offer opt-in programs that gather and share far more information, including device location, the phone numbers you call, and which sites you visit using your mobile and/or home Internet connection. AT&T calls this their Enhanced Relevant Advertising Program; Verizon’s is called Custom Experience Plus.

In 2021, multiple media outlets reported that some Verizon customers were being automatically enrolled in Custom Experience Plus — even after those customers had already opted out of the same program under its previous name — “Verizon Selects.”

If none of the above opt out options work for you, at a minimum you should be able to opt out of CPNI sharing by calling your carrier, or by visiting one of their stores.

THE CASE FOR OPTING OUT

Why should you opt out of sharing CPNI data? For starters, some of the nation’s largest wireless carriers don’t have a great track record in terms of protecting the sensitive information that you give them solely for the purposes of becoming a customer — let alone the information they collect about your use of their services after that point.

In January 2023, T-Mobile disclosed that someone stole data on 37 million customer accounts, including customer name, billing address, email, phone number, date of birth, T-Mobile account number and plan details. In August 2021, T-Mobile acknowledged that hackers made off with the names, dates of birth, Social Security numbers and driver’s license/ID information on more than 40 million current, former or prospective customers who applied for credit with the company.

Last summer, a cybercriminal began selling the names, email addresses, phone numbers, SSNs and dates of birth on 23 million Americans. An exhaustive analysis of the data strongly suggested it all belonged to customers of one AT&T company or another. AT&T stopped short of saying the data wasn’t theirs, but said the records did not appear to have come from its systems and may be tied to a previous data incident at another company.

However frequently the carriers may alert consumers about CPNI breaches, it’s probably nowhere near often enough. Currently, the carriers are required to report a consumer CPNI breach only in cases “when a person, without authorization or exceeding authorization, has intentionally gained access to, used or disclosed CPNI.”

But that definition of breach was crafted eons ago, back when the primary way CPNI was exposed was through “pretexting,” such when the phone company’s employees are tricked into giving away protected customer data.

In January, regulators at the U.S. Federal Communications Commission (FCC) proposed amending the definition of “breach” to include things like inadvertent disclosure — such as when companies expose CPNI data on a poorly-secured server in the cloud. The FCC is accepting public comments on the matter until March 24, 2023.

While it’s true that the leak of CPNI data does not involve sensitive information like Social Security or credit card numbers, one thing AT&T’s breach notice doesn’t mention is that CPNI data — such as balances and payments made — can be abused by fraudsters to make scam emails and text messages more believable when they’re trying to impersonate AT&T and phish AT&T customers.

The other problem with letting companies share or sell your CPNI data is that the wireless carriers can change their privacy policies at any time, and you are assumed to be okay with those changes as long as you keep using their services.

For example, location data from your wireless device is most definitely CPNI, and yet until very recently all of the major carriers sold their customers’ real-time location data to third party data brokers without customer consent.

What was their punishment? In 2020, the FCC proposed fines totaling $208 million against all of the major carriers for selling their customers’ real-time location data. If that sounds like a lot of money, consider that all of the major wireless providers reported tens of billions of dollars in revenue last year (e.g., Verizon’s consumer revenue alone was more than $100 billion last year).

If the United States had federal privacy laws that were at all consumer-friendly and relevant to today’s digital economy, this kind of data collection and sharing would always be opt-in by default. In such a world, the enormously profitable wireless industry would likely be forced to offer clear financial incentives to customers who choose to share this information.

But until that day arrives, understand that the carriers can change their data collection and sharing policies when it suits them. And regardless of whether you actually read any notices about changes to their privacy policies, you will have agreed to those changes as long as you continue using their service.

Who’s Behind the Botnet-Based Service BHProxies?

A security firm has discovered that a six-year-old crafty botnet known as Mylobot appears to be powering a residential proxy service called BHProxies, which offers paying customers the ability to route their web traffic anonymously through compromised computers. Here’s a closer look at Mylobot, and a deep dive into who may be responsible for operating the BHProxies service.

First identified in 2017 by the security firm Deep Instinct, Mylobot employs a number of fairly sophisticated methods to remain undetected on infected hosts, such as running exclusively in the computer’s temporary memory, and waiting 14 days before attempting to contact the botnet’s command and control servers.

Last year, researchers at Minerva Labs spotted the botnet being used to blast out sextortion scams. But according to a new report from BitSight, the Mylobot botnet’s main functionality has always been about transforming the infected system into a proxy.

The Mylobot malware includes more than 1,000 hard-coded and encrypted domain names, any one of which can be registered and used as control networks for the infected hosts. BitSight researchers found significant overlap in the Internet addresses used by those domains and a domain called BHproxies[.]com.

BHProxies sells access to “residential proxy” networks, which allow someone to rent a residential IP address to use as a relay for their Internet communications, providing anonymity and the advantage of being perceived as a residential user surfing the web. The service is currently advertising access to more than 150,000 devices globally.

“At this point, we cannot prove that BHProxies is linked to Mylobot, but we have a strong suspicion,” wrote BitSight’s Stanislas Arnoud.

To test their hypothesis, BitSight obtained 50 proxies from BHProxies. The researchers were able to use 48 of those 50 proxies to browse to a website they controlled — allowing them to record the true IP addresses of each proxy device.

“Among these 48 recovered residential proxies IP addresses, 28 (58.3%) of those were already present in our sinkhole systems, associated with the Mylobot malware family,” Arnoud continued. “This number is probably higher, but we don’t have a full visibility of the botnet. This gave us clear evidence that Mylobot infected computers are used by the BHProxies service.”

BitSight said it is currently seeing more than 50,000 unique Mylobot infected systems every day, and that India appears to be the most targeted country, followed by the United States, Indonesia and Iran.

“We believe we are only seeing part of the full botnet, which may lead to more than 150,000 infected computers as advertised by BHProxies’ operators,” Arnoud wrote.

WHO’S BEHIND BHPROXIES?

The website BHProxies[.]com has been advertised for nearly a decade on the forum Black Hat World by the user BHProxies. BHProxies has authored 129 posts on Black Hat World since 2012, and their last post on the forum was in December 2022.

BHProxies initially was fairly active on Black Hat World between May and November 2012, after which it suddenly ceased all activity. The account didn’t resume posting on the forum until April 2014.

According to cyber intelligence firm Intel 471, the user BHProxies also used the handle “hassan_isabad_subar” and marketed various software tools, including “Subar’s free email creator” and “Subar’s free proxy scraper.”

Intel 471’s data shows that hassan_isabad_subar registered on the forum using the email address jesus.fn.christ@gmail.com. In a June 2012 private message exchange with a website developer on Black Hat World, hassan_isabad_subar confided that they were working at the time to develop two websites, including the now-defunct customscrabblejewelry.com.

DomainTools.com reports that customscrabblejewelry.com was registered in 2012 to a Teresa Shotliff in Chesterland, Ohio. A search on jesus.fn.christ@gmail.com at Constella Intelligence, a company that tracks compromised databases, shows this email address is tied to an account at the fundraising platform omaze.com, for a Brian Shotliff from Chesterland, Ohio.

Reached via LinkedIn, Mr. Shotliff said he sold his BHProxies account to another Black Hat World forum user from Egypt back in 2014. Shotliff shared an April 2014 password reset email from Black Hat World, which shows he forwarded the plaintext password to the email address legendboy2050@yahoo.com. He also shared a PayPal receipt and snippets of Facebook Messenger logs showing conversations in March 2014 with legendboy2050@yahoo.com.

Constella Intelligence confirmed that legendboy2050@yahoo.com was indeed another email address tied to the hassan_isabad_subar/BHProxies identity on Black Hat World. Constella also connects legendboy2050 to Facebook and Instagram accounts for one Abdala Tawfik from Cairo. This user’s Facebook page says Tawfik also uses the name Abdalla Khafagy.

Tawfik’s Instagram account says he is a former operations manager at the social media network TikTok, as well as a former director at Crypto.com.

Abdalla Khafagy’s LinkedIn profile says he was “global director of community” at Crypto.com for about a year ending in January 2022. Before that, the resume says he was operations manager of TikTok’s Middle East and North Africa region for approximately seven months ending in April 2020.

Khafagy’s LinkedIn profile says he is currently founder of LewkLabs, a Dubai-based “blockchain-powered, SocialFi content monetization platform” that last year reported funding of $3.26 million from private investors.

The only experience listed for Khafagy prior to the TikTok job is labeled “Marketing” at “Confidential,” from February 2014 to October 2019.

Reached via LinkedIn, Mr. Khafagy told KrebsOnSecurity that he had a Black Hat World account at some point, but that he didn’t recall ever having used an account by the name BHProxies or hassan_isabad_subar. Khafagy said he couldn’t remember the name of the account he had on the forum.

“I had an account that was simply hacked from me shortly after and I never bothered about it because it wasn’t mine in the first place,” he explained.

Khafagy declined to elaborate on the five-year stint in his resume marked “Confidential.” When asked directly whether he had ever been associated with the BHProxies service, Mr. Khafagy said no.

That Confidential job listing is interesting because its start date lines up with the creation of BHproxies[.]com. Archive.org indexed its first copy of BHProxies[.]com on Mar. 5, 2014, but historic DNS records show BHproxies[.]com first came online Feb. 25, 2014.

Shortly after that conversation with Mr. Khafagy, Mr. Shotliff shared a Facebook/Meta message he received that indicated Mr. Khafagy wanted him to support the claim that the BHProxies account had somehow gone missing.

“Hey mate, it’s been a long time. Hope you are doing well. Someone from Krebs on Security reached out to me about the account I got from you on BHW,” Khafagy’s Meta account wrote. “Didn’t we try to retrieve this account? I remember mentioning to you that it got stolen and I was never able to retrieve it.”

Mr. Shotliff said Khafagy’s sudden message this week was the first time he’d heard that claim.

“He bought the account,” Shotliff said. “He might have lost the account or had it stolen, but it’s not something I remember.”

If you liked this story, you may also enjoy these other investigations into botnet-based proxy services:

A Deep Dive Into the Residential Proxy Service ‘911’
911 Proxy Service Implodes After Disclosing Breach
Meet the Administrators of the RSOCKS Proxy Botnet
The Link Between AWM Proxy & the Glupteba Botnet
15-Year-Old Malware Proxy Network VIP72 Goes Dark
Who’s Behind the TDSS Botnet?

New Protections for Food Benefits Stolen by Skimmers

Millions of Americans receiving food assistance benefits just earned a new right that they can’t yet enforce: The right to be reimbursed if funds on their Electronic Benefit Transfer (EBT) cards are stolen by card skimming devices secretly installed at cash machines and grocery store checkout lanes.

On December 29, 2022, President Biden signed into law the Consolidated Appropriations Act of 2023, which — for the first time ever — includes provisions for the replacement of stolen EBT benefits. This is a big deal because in 2022, organized crime groups began massively targeting EBT accounts — often emptying affected accounts at ATMs immediately after the states disperse funds each month.

EBT cards can be used along with a personal identification number (PIN) to pay for goods at participating stores, and to withdraw cash from an ATM. However, EBT cards differ from debit cards issued to most Americans in two important ways. First, most states do not equip EBT cards with smart chip technology, which can make the cards more difficult and expensive for skimming thieves to clone.

More critically, EBT participants traditionally have had little hope of recovering food assistance funds when their cards were copied by card-skimming devices and used for fraud. That’s because while the EBT programs are operated by individually by the states, those programs are funded by the U.S. Department of Agriculture (USDA), which until late last year was barred from reimbursing states for stolen EBT funds.

The protections passed in the 2023 Appropriations Act allow states to use federal funds to replace stolen EBT benefits, and they permit states to seek reimbursement for any skimmed EBT funds they may have replaced from their own coffers (dating back to Oct. 1, 2022).

But first, all 50 states must each submit a plan for how they are going to protect and replace food benefits stolen via card skimming. Guidance for the states in drafting those plans was issued by the USDA on Jan. 31 (PDF), and states that don’t get them done before Feb. 27, 2023 risk losing the ability to be reimbursed for EBT fraud losses.

Deborah Harris is a staff attorney at The Massachusetts Law Reform Institute (MLRI), a nonprofit legal assistance organization that has closely tracked the EBT skimming epidemic. In November 2022, the MLRI filed a class-action lawsuit against Massachusetts on behalf of thousands of low-income families who were collectively robbed of more than $1 million in food assistance benefits by card skimming devices secretly installed at cash machines and grocery store checkout lanes across the state.

Harris said she’s pleased that the USDA guidelines were issued so promptly, and that the guidance for states was not overly prescriptive. For example, some security experts have suggested that adding contactless capability to EBT cards could help participants avoid skimming devices altogether. But Harris said contactless cards do not require a PIN, which is the only thing that stops EBT cards from being drained at the ATM when a participant’s card is lost or stolen.

Then again, nothing in the guidance even mentions chip-based cards, or any other advice for improving the physical security of EBT cards. Rather, it suggests states should seek to develop the capability to perform basic fraud detection and alerting on suspicious transactions, such as when an EBT card that is normally used only in one geographic area suddenly is used to withdraw cash at an ATM halfway across the country.

“Besides having the states move fast to approve their plans, we’d also like to see a focused effort to move states from magstripe-only cards to chip, and also assisting states to develop the algorithms that will enable them to identify likely incidents of stolen benefits,” Harris said.

Harris said Massachusetts has begun using algorithms to look for these suspicious transaction patterns throughout its EBT network, and now has the ability to alert households and verify transactions. But she said most states do not have this capability.

“We have heard that other states aren’t currently able to do that,” Harris said. “But encouraging states to more affirmatively identify instances of likely theft and assisting with the claims and verification process is critical. Most households can’t do that on their own, and in Massachusetts it’s very hard for a person to get a copy of their transaction history. Some states can do that through third-party apps, but something so basic should not be on the burden of EBT households.”

Some states aren’t waiting for direction from the federal government to beef up EBT card security. Like Maryland, which identified more than 1,400 households hit by EBT skimming attacks last year — a tenfold increase over 2021.

Advocates for EBT beneficiaries in Maryland are backing Senate Bill 401 (PDF), which would require the use of chip technology and ongoing monitoring for suspicious activity (a hearing on SB401 is scheduled in the Maryland Senate Finance Commission for Thursday, Feb. 23, at 1 p.m.).

Michelle Salomon Madaio is a director at the Homeless Persons Representation Project, a legal assistance organization based in Silver Spring, Md. Madaio said the bill would require the state Department of Human Services to replace skimmed benefits, not only after the bill goes into effect but also retroactively from January 2020 to the present.

Madaio said the bill also would require the state to monitor for patterns of suspicious activity on EBT cards, and to develop a mechanism to contact potentially affected households.

“For most of the skimming victims we’ve worked with, the fraudulent transactions would be pretty easy to spot because they mostly happened in the middle of the night or out of state, or both,” Madaio said. “To make matters worse, a lot of families whose benefits were scammed then incurred late fees on many other things as a result.”

It is not difficult to see why organized crime groups have pounced on EBT cards as easy money. In most traditional payment card transactions, there are usually several parties that have a financial interest in minimizing fraud and fraud losses, including the bank that issued the card, the card network (Visa, MasterCard, Discover, etc.), and the merchant.

But that infrastructure simply does not exist within state EBT programs, and it certainly isn’t a thing at the inter-state level. What that means is that the vast majority of EBT cards have zero fraud controls, which is exactly what continues to make them so appealing to thieves.

For now, the only fraud controls available to most EBT cardholders include being especially paranoid about where they use their cards, and frequently changing their PINs.

According to USDA guidance issued prior to the passage of the appropriations act, EBT cardholders should consider changing their card PIN at least once a month.

“By changing PINs frequently, at least monthly, and doing so before benefit issuance dates, households can minimize their risk of stolen benefits from a previously skimmed EBT card,” the USDA advised.

Administrator of RSOCKS Proxy Botnet Pleads Guilty

Denis Emelyantsev, a 36-year-old Russian man accused of running a massive botnet called RSOCKS that stitched malware into millions of devices worldwide, pleaded guilty to two counts of computer crime violations in a California courtroom this week. The plea comes just months after Emelyantsev was extradited from Bulgaria, where he told investigators, “America is looking for me because I have enormous information and they need it.”

First advertised in the cybercrime underground in 2014, RSOCKS was the web-based storefront for hacked computers that were sold as “proxies” to cybercriminals looking for ways to route their Web traffic through someone else’s device.

Customers could pay to rent access to a pool of proxies for a specified period, with costs ranging from $30 per day for access to 2,000 proxies, to $200 daily for up to 90,000 proxies.

Many of the infected systems were Internet of Things (IoT) devices, including industrial control systems, time clocks, routers, audio/video streaming devices, and smart garage door openers. Later in its existence, the RSOCKS botnet expanded into compromising Android devices and conventional computers.

In June 2022, authorities in the United States, Germany, the Netherlands and the United Kingdom announced a joint operation to dismantle the RSOCKS botnet. But that action did not name any defendants.

Inspired by that takedown, KrebsOnSecurity followed clues from the RSOCKS botnet master’s identity on the cybercrime forums to Emelyantsev’s personal blog, where he went by the name Denis Kloster. The blog featured musings on the challenges of running a company that sells “security and anonymity services to customers around the world,” and even included a group photo of RSOCKS employees.

“Thanks to you, we are now developing in the field of information security and anonymity!,” Kloster’s blog enthused. “We make products that are used by thousands of people around the world, and this is very cool! And this is just the beginning!!! We don’t just work together and we’re not just friends, we’re Family.”

But by the time that investigation was published, Emelyantsev had already been captured by Bulgarian authorities responding to an American arrest warrant. At his extradition hearing, Emelyantsev claimed he would prove his innocence in an U.S. courtroom.

“I have hired a lawyer there and I want you to send me as quickly as possible to clear these baseless charges,” Emelyantsev told the Bulgarian court. “I am not a criminal and I will prove it in an American court.”

Emelyantsev was far more than just an administrator of a large botnet. Behind the facade of his Internet advertising company based in Omsk, Russia, the RSOCKS botmaster was a major player in the Russian email spam industry for more than a decade.

Some of the top Russian cybercrime forums have been hacked over the years, and leaked private messages from those forums show the RSOCKS administrator claimed ownership of the RUSdot spam forum. RUSdot is the successor forum to Spamdot, a far more secretive and restricted community where most of the world’s top spammers, virus writers and cybercriminals collaborated for years before the forum imploded in 2010.

Indeed, the very first mentions of RSOCKS on any Russian-language cybercrime forums refer to the service by its full name as the “RUSdot Socks Server.”

Email spam — and in particular malicious email sent via compromised computers — is still one of the biggest sources of malware infections that lead to data breaches and ransomware attacks. So it stands to reason that as administrator of Russia’s most well-known forum for spammers, Emelyantsev probably knows quite a bit about other top players in the botnet spam and malware community.

It remains unclear whether Emelyantsev made good on his promise to spill that knowledge to American investigators as part of his plea deal. The case is being prosecuted by the U.S. Attorney’s Office for the Southern District of California, which has not responded to a request for comment.

Emelyantsev pleaded guilty on Monday to two counts, including damage to protected computers and conspiracy to damage protected computers. He faces a maximum of 20 years in prison, and is currently scheduled to be sentenced on April 27, 2023.

Lawsuit Seeks Food Benefits Stolen By Skimmers

A nonprofit organization is suing the state of Massachusetts on behalf of thousands of low-income families who were collectively robbed of more than a $1 million in food assistance benefits by card skimming devices secretly installed at cash machines and grocery store checkout lanes across the state. Federal law bars states from replacing these benefits using federal funds, and a recent rash of skimming incidents nationwide has disproportionately affected those receiving food assistance via state-issued prepaid debit cards.

On Nov. 4, The Massachusetts Law Reform Institute (MLRI) filed a class action lawsuit on behalf of low-income families whose Supplemental Nutrition and Assistance Program (SNAP) benefits were stolen from their accounts. The SNAP program serves over a million people in Massachusetts, and 41 million people nationally.

“Over the past few months, thieves have stolen over a million SNAP dollars from thousands of Massachusetts families – putting their nutrition and economic stability at risk,” the MLRI said in a statement on the lawsuit. “The criminals attach a skimming device on a POS (point of sale) terminal to capture the household’s account information and PIN. The criminals then use that information to make a fake card and steal the SNAP benefits.”

In announcing the lawsuit, the MRLI linked to a story KrebsOnSecurity published last month that examined how skimming thieves increasingly are targeting SNAP payment card holders nationwide. The story looked at how the vast majority of SNAP benefit cards issued by the states do not include the latest chip technology that makes it more difficult and expensive for thieves to clone them.

The story also highlighted how SNAP cardholders usually have little recourse to recover any stolen funds — even in unlikely cases where the victim has gathered mountains of proof to show state and federal officials that the fraudulent withdrawals were not theirs.

Deborah Harris is a staff attorney at the MLRI. Harris said the goal of the lawsuit is to force Massachusetts to reimburse SNAP skimming victims using state funds, and to convince The U.S. Department of Agriculture (USDA) — which funds the program that states draw from — to change its policies and allow states to replace stolen benefits with federal funds.

“Ultimately we think it’s the USDA that needs to step up and tell states they have a duty to restore the stolen benefits, and that USDA will cover the cost at least until there is better security in place, such as chip cards,” Harris told KrebsOnSecurity.

“The losses we’re talking about are relatively small in the scheme of total SNAP expenditures which are billions,” she said. “But if you are a family that can’t pay for food because you suddenly don’t have money in your account, it’s devastating for the family.”

The USDA has not said it will help states restore the stolen funds. But on Oct. 31, 2022, the agency released guidance (PDF) whose primary instructions were included in an appendix titled, Card Security Options Available to Households. Notably, the USDA did not mention the idea of shifting to chip-based SNAP benefits cards.

“The guidance generally continues to make households responsible for preventing the theft of their benefits as well as for suffering the loss when benefits are stolen through no fault of the household,” Harris said. “Many of the recommendations are not practical for households who don’t have a smartphone to receive text messages and aren’t able to change their PIN after each transaction and keep track of the new PIN.”

Harris said three of the four recommendations are not currently available in Massachusetts, and they are very likely not currently available in other states. For example, she said, Massachusetts households do not have the option of freezing or locking their cards between transactions. Nor do they receive alerts about transactions. And they most certainly don’t have any way to block out-of-state transactions.

“Perhaps these are options that [card] processors and states could provide, but they are not available now as far as we know,” Harris said. “Most likely they would take time to implement.”

Accused Russian RSOCKS Botmaster Arrested, Requests Extradition to U.S.

A 36-year-old Russian man recently identified by KrebsOnSecurity as the likely proprietor of the massive RSOCKS botnet has been arrested in Bulgaria at the request of U.S. authorities. At a court hearing in Bulgaria this month, the accused hacker requested and was granted extradition to the United States, reportedly telling the judge, “America is looking for me because I have enormous information and they need it.”

On June 22, KrebsOnSecurity published Meet the Administrators of the RSOCKS Proxy Botnet, which identified Denis Kloster, a.k.a. Denis Emelyantsev, as the apparent owner of RSOCKS, a collection of millions of hacked devices that were sold as “proxies” to cybercriminals looking for ways to route their malicious traffic through someone else’s computer.

A native of Omsk, Russia, Kloster came into focus after KrebsOnSecurity followed clues from the RSOCKS botnet master’s identity on the cybercrime forums to Kloster’s personal blog, which featured musings on the challenges of running a company that sells “security and anonymity services to customers around the world.” Kloster’s blog even included a group photo of RSOCKS employees.

“Thanks to you, we are now developing in the field of information security and anonymity!,” Kloster’s blog enthused. “We make products that are used by thousands of people around the world, and this is very cool! And this is just the beginning!!! We don’t just work together and we’re not just friends, we’re Family.”

The Bulgarian news outlet 24Chasa.bg reports that Kloster was arrested in June at a co-working space in the southwestern ski resort town of Bansko, and that the accused asked to be handed over to the American authorities.

“I have hired a lawyer there and I want you to send me as quickly as possible to clear these baseless charges,” Kloster reportedly told the Bulgarian court this week. “I am not a criminal and I will prove it in an American court.”

Launched in 2013, RSOCKS was shut down in June 2022 as part of an international investigation into the cybercrime service. According to the Justice Department, the RSOCKS botnet initially targeted Internet of Things (IoT) devices, including industrial control systems, time clocks, routers, audio/video streaming devices, and smart garage door openers; later in its existence, the RSOCKS botnet expanded into compromising additional types of devices, including Android devices and conventional computers, the DOJ said.

The Justice Department’s June 2022 statement about that takedown cited a search warrant from the U.S. Attorney’s Office for the Southern District of California, which also was named by Bulgarian news outlets this month as the source of Kloster’s arrest warrant.

When asked about the existence of an arrest warrant or criminal charges against Kloster, a spokesperson for the Southern District said, “no comment.”

Update, Sept. 24, 9:00 a.m. ET: Kloster was named in a 2019 indictment (PDF) unsealed Sept. 23 by the Southern District court.

24Chasa said the defendant’s surname is Emelyantsev and that he only recently adopted the last name Kloster, which is his mother’s maiden name.

As KrebsOnSecurity reported in June, Kloster also appears to be a major player in the Russian email spam industry. In several private exchanges on cybercrime forums, the RSOCKS administrator claimed ownership of the RUSdot spam forum. RUSdot is the successor forum to Spamdot, a far more secretive and restricted forum where most of the world’s top spammers, virus writers and cybercriminals collaborated for years before the community’s implosion in 2010.

Email spam — and in particular malicious email sent via compromised computers — is still one of the biggest sources of malware infections that lead to data breaches and ransomware attacks. So it stands to reason that as administrator of Russia’s most well-known forum for spammers, the defendant in this case probably knows quite a bit about other top players in the botnet spam and malware community.

Despite maintaining his innocence, Kloster reportedly told the Bulgarian judge that he could be useful to American investigators.

“America is looking for me because I have enormous information and they need it,” Kloster told the court, according to 24Chasa. “That’s why they want me.”

The Bulgarian court agreed, and granted his extradition. Kloster’s fiancee also attended the extradition hearing, and reportedly wept in the hall outside the entire time.

Kloster turned 36 while awaiting his extradition hearing, and may soon be facing charges that carry punishments of up to 20 years in prison.

No SOCKS, No Shoes, No Malware Proxy Services!

With the recent demise of several popular “proxy” services that let cybercriminals route their malicious traffic through hacked PCs, there is now something of a supply chain crisis gripping the underbelly of the Internet. Compounding the problem, several remaining malware-based proxy services have chosen to block new registrations to avoid swamping their networks with a sudden influx of customers.

Last week, a seven-year-old proxy service called 911[.]re abruptly announced it was permanently closing after a cybersecurity breach allowed unknown intruders to trash its servers and delete customer data and backups. 911 was already akin to critical infrastructure for many in the cybercriminal community after its top two competitors — VIP72 and LuxSocks — closed or were shut down by authorities over the past 10 months.

The underground cybercrime forums are now awash in pleas from people who are desperately seeking a new supplier of abundant, cheap, and reliably clean proxies to restart their businesses. The consensus seems to be that those days are now over, and while there are many smaller proxy services remaining, few of them on their own are capable of absorbing anywhere near the current demand.

“Everybody is looking for an alternative, bro,” wrote a BlackHatForums user on Aug. 1 in response to one of many “911 alternative” discussion threads. “No one knows an equivalent alternative to 911[.]re. Their service in terms of value and accessibility compared to other proxy providers was unmatched. Hopefully someone comes with a great alternative to 911[.]re.”

Among the more frequently recommended alternatives to 911 is SocksEscort[.]com, a malware-based proxy network that has been in existence since at least 2010. Here’s what part of their current homepage looks like:

But faced with a deluge of new signups in the wake of 911’s implosion, SocksEscort was among the remaining veteran proxy services that opted to close its doors to new registrants, replacing its registration page with the message:

“Due to unusual high demand, and heavy load on our servers, we had to block all new registrations. We won’t be able to support our proxies otherwise, and close SocksEscort as a result. We will resume registrations right after demand drops. Thank you for understanding, and sorry for the inconvenience.”

According to Spur.us, a startup that tracks proxy services, SocksEscort is a malware-based proxy offering, which means the machines doing the proxying of traffic for SocksEscort customers have been infected with malicious software that turns them into a traffic relay.

Spur says SocksEscort’s proxy service relies on software designed to run on Windows computers, and is currently leasing access to more than 14,000 hacked computers worldwide. That is a far cry from the proxy inventory advertised by 911, which stood at more than 200,000 IP addresses for rent just a few days ago.

SocksEscort is what’s known as a “SOCKS Proxy” service. The SOCKS (or SOCKS5) protocol allows Internet users to channel their Web traffic through a proxy server, which then passes the information on to the intended destination. From a website’s perspective, the traffic of the proxy network customer appears to originate from a rented/malware-infected PC tied to a residential ISP customer, not from the proxy service customer.

These services can be used in a legitimate manner for several business purposes — such as price comparisons or sales intelligence — but they are massively abused for hiding cybercrime activity because they make it difficult to trace malicious traffic to its original source.

The disruption at 911[.]re came days after KrebsOnSecurity published an in-depth look at the long-running proxy service, which showed that 911 had a history of incentivizing the installation of its proxy software without user notice or consent, and that it actually ran some of these “pay-per-install” schemes on its own to guarantee a steady supply of freshly-hacked PCs.

More on SocksEscort in an upcoming story.

Further reading:

July 29, 2022: 911 Proxy Service Implodes After Disclosing Breach

July 28, 2022: Breach Exposes Users of Microleaves Proxy Service

July 18, 2022: A Deep Dive Into the Residential Proxy Service ‘911’

June 28, 2022: The Link Between AWM Proxy & the Glupteba Botnet

June 22, 2022: Meet the Administrators of the RSOCKS Proxy Botnet

Sept. 1, 2021: 15-Year-Old Malware Proxy Network VIP72 Goes Dark

911 Proxy Service Implodes After Disclosing Breach

911[.]re, a proxy service that since 2015 has sold access to hundreds of thousands of Microsoft Windows computers daily, announced this week that it is shutting down in the wake of a data breach that destroyed key components of its business operations. The abrupt closure comes ten days after KrebsOnSecurity published an in-depth look at 911 and its connections to shady pay-per-install affiliate programs that secretly bundled 911’s proxy software with other titles, including “free” utilities and pirated software.

911[.]re is was one of the original “residential proxy” networks, which allow someone to rent a residential IP address to use as a relay for his/her Internet communications, providing anonymity and the advantage of being perceived as a residential user surfing the web.

Residential proxy services are often marketed to people seeking the ability to evade country-specific blocking by the major movie and media streaming providers. But some of them — like 911 — build their networks in part by offering “free VPN” or “free proxy” services that are powered by software which turns the user’s PC into a traffic relay for other users. In this scenario, users indeed get to use a free VPN service, but they are often unaware that doing so will turn their computer into a proxy that lets others use their Internet address to transact online.

From a website’s perspective, the IP traffic of a residential proxy network user appears to originate from the rented residential IP address, not from the proxy service customer. These services can be used in a legitimate manner for several business purposes — such as price comparisons or sales intelligence — but they are massively abused for hiding cybercrime activity because they can make it difficult to trace malicious traffic to its original source.

As noted in KrebsOnSecurity’s July 19 story on 911, the proxy service operated multiple pay-per-install schemes that paid affiliates to surreptitiously bundle the proxy software with other software, continuously generating a steady stream of new proxies for the service.

Within hours of that story, 911 posted a notice at the top of its site, saying, “We are reviewing our network and adding a series of security measures to prevent misuse of our services. Proxy balance top-up and new user registration are closed. We are reviewing every existing user, to ensure their usage is legit and [in] compliance with our Terms of Service.”

At this announcement, all hell broke loose on various cybercrime forums, where many longtime 911 customers reported they were unable to use the service. Others affected by the outage said it seemed 911 was trying to implement some sort of “know your customer” rules — that maybe 911 was just trying to weed out those customers using the service for high volumes of cybercriminal activity.

Then on July 28, the 911 website began redirecting to a notice saying, “We regret to inform you that we permanently shut down 911 and all its services on July 28th.”

According to 911, the service was hacked in early July, and it was discovered that someone manipulated the balances of a large number of user accounts. 911 said the intruders abused an application programming interface (API) that handles the topping up of accounts when users make financial deposits with the service.

“Not sure how did the hacker get in,” the 911 message reads. “Therefore, we urgently shut down the recharge system, new user registration, and an investigation started.”

However the intruders got in, 911 said, they managed to also overwrite critical 911[.]re servers, data and backups of that data.

“On July 28th, a large number of users reported that they could not log in the system,” the statement continues. “We found that the data on the server was maliciously damaged by the hacker, resulting in the loss of data and backups. Its [sic] confirmed that the recharge system was also hacked the same way. We were forced to make this difficult decision due to the loss of important data that made the service unrecoverable.”

Operated largely out of China, 911 was an enormously popular service across many cybercrime forums, and it became something akin to critical infrastructure for this community after two of 911’s longtime competitors — malware-based proxy services VIP72 and LuxSocks — closed their doors in the past year.

Now, many on the crime forums who relied on 911 for their operations are wondering aloud whether there are any alternatives that match the scale and utility that 911 offered. The consensus seems to be a resounding “no.”

I’m guessing we may soon learn more about the security incidents that caused 911 to implode. And perhaps other proxy services will spring up to meet what appears to be a burgeoning demand for such services at the moment, with comparatively little supply.

In the meantime, 911’s absence may coincide with a measurable (if only short-lived) reprieve in unwanted traffic to top Internet destinations, including banks, retailers and cryptocurrency platforms, as many former customers of the proxy service scramble to make alternative arrangements.

Riley Kilmer, co-founder of the proxy-tracking service Spur.us, said 911’s network will be difficult to replicate in the short run.

“My speculation is [911’s remaining competitors] are going to get a major boost in the short term, but a new player will eventually come along,” Kilmer said. “None of those are good replacements for LuxSocks or 911. However, they will all allow anyone to use them. For fraud rates, the attempts will continue but through these replacement services which should be easier to monitor and stop. 911 had some very clean IP addresses.”

911 wasn’t the only major proxy provider disclosing a breach this week tied to unauthenticated APIs: On July 28, KrebsOnSecurity reported that internal APIs exposed to the web had leaked the customer database for Microleaves, a proxy service that rotates its customers’ IP addresses every five to ten minutes. That investigation showed Microleaves — like 911 — had a long history of using pay-per-install schemes to spread its proxy software.

Breach Exposes Users of Microleaves Proxy Service

Microleaves, a ten-year-old proxy service that lets customers route their web traffic through millions of Microsoft Windows computers, recently fixed a vulnerability in their website that exposed their entire user database. Microleaves claims its proxy software is installed with user consent, but data exposed in the breach shows the service has a lengthy history of being supplied with new proxies by affiliates incentivized to distribute the software any which way they can — such as by secretly bundling it with other titles.

Launched in 2013, Microleaves is a service that allows customers to route their Internet traffic through PCs in virtually any country or city around the globe. Microleaves works by changing each customer’s Internet Protocol (IP) address every five to ten minutes.

The service, which accepts PayPal, Bitcoin and all major credit cards, is aimed primarily at enterprises engaged in repetitive, automated activity that often results in an IP address being temporarily blocked — such as data scraping, or mass-creating new accounts at some service online.

In response to a report about the data exposure from KrebsOnSecurity, Microleaves said it was grateful for being notified about a “very serious issue regarding our customer information.”

Abhishek Gupta is the PR and marketing manager for Microleaves, which he said in the process of being rebranded to “Shifter.io.” Gupta said the report qualified as a “medium” severity security issue in Shifter’s brand new bug bounty program (the site makes no mention of a bug bounty), which he said offers up to $2,000 for reporting data exposure issues like the one they just fixed. KrebsOnSecurity declined the offer and requested that Shifter donate the amount to the Electronic Frontier Foundation (EFF), a digital rights group.

From its inception nearly a decade ago, Microleaves has claimed to lease between 20-30 million IPs via its service at any time. Riley Kilmer, co-founder of the proxy-tracking service Spur.us, said that 20-30 million number might be accurate for Shifter if measured across a six-month time frame. Currently, Spur is tracking roughly a quarter-million proxies associated with Microleaves/Shifter each day, with a high rate of churn in IPs.

Early on, this rather large volume of IP addresses led many to speculate that Microleaves was just a botnet which was being resold as a commercial proxy service.

The very first discussion thread started by the new user Microleaves on the forum BlackHatWorld in 2013 sought forum members who could help test and grow the proxy network. At the time, the Microleaves user said their proxy network had 150,000 IPs globally, and was growing quickly.

One of BlackHatWorld’s moderators asked the administrator of the forum to review the Microleaves post.

“User states has 150k proxies,” the forum skeptic wrote. “No seller on BHW has 150k working daily proxies none of us do. Which hints at a possible BOTNET. That’s the only way you will get 150k.”

Microleaves has long been classified by antivirus companies as adware or as a “potentially unwanted program” (PUP), the euphemism that antivirus companies use to describe executable files that get installed with ambiguous consent at best, and are often part of a bundle of software tied to some “free” download. Security vendor Kaspersky flags the Microleaves family of software as a trojan horse program that commandeers the user’s Internet connection as a proxy without notifying the user.

“While working, these Trojans pose as Microsoft Windows Update,” Kaspersky wrote.

In a February 2014 post to BlackHatWorld, Microleaves announced that its sister service — reverseproxies[.]com — was now offering an “Auto CAPTCHA Solving Service,” which automates the solving of those squiggly and sometimes frustrating puzzles that many websites use to distinguish bots from real visitors. The CAPTCHA service was offered as an add-on to the Microleaves proxy service, and ranged in price from $20 for a 2-day trial to $320 for solving up to 80 captchas simultaneously.

“We break normal Recaptcha with 60-90% success rate, recaptcha with blobs 30% success, and 500+ other captcha,” Microleaves wrote. “As you know all success rate on recaptcha depends very much on good proxies that are fresh and not spammed!”

WHO IS ACIDUT?

The exposed Microleaves user database shows that the first user created on the service — username “admin” — used the email address alex.iulian@aol.com. A search on that email address in Constella Intelligence, a service that tracks breached data, reveals it was used to create an account at the link shortening service bit.ly under the name Alexandru Florea, and the username “Acidut.” [Full disclosure: Constella is currently an advertiser on this website].

According to the cyber intelligence company Intel 471, a user named Acidut with the email address iulyan87_4u@gmail.com had an active presence on almost a dozen shadowy money-making and cybercrime forums from 2010 to 2017, including BlackHatWorld, Carder[.]pro, Hackforums, OpenSC, and CPAElites.

In a 2011 post on Hackforums, Acidut said they were building a botnet using an “exploit kit,” a set of browser exploits made to be stitched into hacked websites and foist malware on visitors. Acidut claimed their exploit kit was generating 3,000 to 5,000 new bots each day. OpenSC was hacked at one point, and its private messages show Acidut purchased a license from Exmanoize, the handle used by the creator of the Eleonore Exploit Kit.

By November 2013, Acidut was advertising the sale of “26 million SOCKS residential proxies.” In a March 2016 post to CPAElites, Acidut said they had a worthwhile offer for people involved in pay-per-install or “PPI” schemes, which match criminal gangs who pay for malware installs with enterprising hackers looking to sell access to compromised PCs and websites.

Because pay-per-install affiliate schemes rarely impose restrictions on how the software can be installed, such programs can be appealing for cybercriminals who already control large collections of hacked machines and/or compromised websites. Indeed, Acidut went a step further, adding that their program could be quietly and invisibly nested inside of other programs.

“For those of you who are doing PPI I have a global offer that you can bundle to your installer,” Acidut wrote. “I am looking for many installs for an app that will generate website visits. The installer has a silence version which you can use inside your installer. I am looking to buy as many daily installs as possible worldwide, except China.”

Asked about the source of their proxies in 2014, the Microleaves user responded that it was “something related to a PPI network. I can’t say more and I won’t get into details.”

Acidut authored a similar message on the forum BlackHatWorld in 2013, where they encouraged users to contact them on Skype at the username “nevo.julian.” That same Skype contact address was listed prominently on the Microleaves homepage up until about a week ago when KrebsOnSecurity first reached out to the company.

ONLINE[.]IO (NOW MERCIFULLY OFFLINE)

There is a Facebook profile for an Alexandru Iulian Florea from Constanta, Romania, whose username on the social media network is Acidut. Prior to KrebsOnSecurity alerting Shifter of its data breach, the Acidut profile page associated Florea with the websites microleaves.com, shrooms.io, leftclick[.]io, and online[.]io. Mr. Florea did not respond to multiple requests for comment, and his Facebook page no longer mentions these domains.

Leftclick and online[.]io emerged as subsidiaries of Microleaves between 2017 and 2018. According to a help wanted ad posted in 2018 for a developer position at online[.]io, the company’s services were brazenly pitched to investors as “a cybersecurity and privacy tool kit, offering extensive protection using advanced adblocking, anti-tracking systems, malware protection, and revolutionary VPN access based on residential IPs.”

“Online[.]io is developing the first fully decentralized peer-to-peer networking technology and revolutionizing the browsing experience by making it faster, ad free, more reliable, secure and non-trackable, thus freeing the Internet from annoying ads, malware, and trackers,” reads the rest of that help wanted ad.

Microleaves CEO Alexandru Florea gave an “interview” to the website Irishtechnews.ie in 2018, in which he explained how Online[.]io (OIO) was going to upend the online advertising and security industries with its initial coin offering (ICO). The word interview is in air quotes because the following statements by Florea deserved some serious pushback by the interviewer.

“Online[.]io solution, developed using the Ethereum blockchain, aims at disrupting the digital advertising market valued at more than $1 trillion USD,” Alexandru enthused. “By staking OIO tokens and implementing our solution, the website operators will be able to access a new non-invasive revenue stream, which capitalizes on time spent by users online.”

“At the same time, internet users who stake OIO tokens will have the opportunity to monetize on the time spent online by themselves and their peers on the World Wide Web,” he continued. “The time spent by users online will lead to ICE tokens being mined, which in turn can be used in the dedicated merchant system or traded on exchanges and consequently changed to fiat.”

Translation: If you install our proxy bot/CAPTCHA-solver/ad software on your computer — or as an exploit kit on your website — we’ll make millions hijacking ads and you will be rewarded with heaps of soon-to-be-worthless shitcoin. Oh, and all your security woes will disappear, too.

It’s unclear how many Internet users and websites willingly agreed to get bombarded with Online[.]io’s annoying ads and search hijackers — and to have their PC turned into a proxy or CAPTCHA-solving zombie for others. But that is exactly what multiple security companies said happened when users encountered online[.]io, which operated using the Microsoft Windows process name of “online-guardian.exe.”

Incredibly, Crunchbase says Online[.]io raised $6 million in funding for an initial coin offering in 2018, based on the plainly ludicrous claims made above. Since then, however, online[.]io seems to have gone…offline, for good.

SUPER TECH VENTURES?

Until this week, Shifter.io’s website also exposed information about its customer base and most active users, as well as how much money each client has paid over the lifetime of their subscription. The data indicates Shifter has earned more than $11.7 million in direct payments, although it’s unclear how far back in time those payment records go, or how complete they are.

The bulk of Shifter customers who spent more than $100,000 at the proxy service appear to be digital advertising companies, including some located in the United States. None of the several Shifter customers approached by KrebsOnSecurity agreed to be interviewed.

Shifter’s Gupta said he’d been with the company for three years, since the new owner took over the company and made the rebrand to Shifter.

“The company has been on the market for a long time, but operated under a different brand called Microleaves, until new ownership and management took over the company started a reorganization process that is still on-going,” Gupta said. “We are fully transparent. Mostly [our customers] work in the data scraping niche, this is why we actually developed more products in this zone and made a big shift towards APIs and integrated solutions in the past year.”

Ah yes, the same APIs and integrated solutions that were found exposed to the Internet and leaking all of Shifter’s customer information.

Gupta said the original founder of Microleaves was a man from India, who later sold the business to Florea. According to Gupta, the Romanian entrepreneur had multiple issues in trying to run the company, and then sold it three years ago to the current owner — Super Tech Ventures, a private equity company based in Taiwan.

“Our CEO is Wang Wei, he has been with the company since 3 years ago,” Gupta said. “Mr. Florea left the company two years ago after ending this transition period.”

Google and other search engines seem to know nothing about a Super Tech Ventures based in Taiwan. Incredibly, Shifter’s own PR person claimed that he, too, was in the dark on this subject.

“I would love to help, but I really don’t know much about the mother company,” Gupta said, essentially walking back his “fully transparent” statement. “I know they are a branch of the bigger group of asian investment firms focused on private equity in multiple industries.”

Adware and proxy software are often bundled together with “free” software utilities online, or with popular software titles that have been pirated and quietly fused with installers tied to various PPI affiliate schemes.

But just as often, these intrusive programs will include some type of notice — even if installed as part of a software bundle — that many users simply do not read and click “Next” to get on with installing whatever software they’re seeking to use. In these cases, selecting the “basic” or “default” settings while installing usually hides any per-program installation prompts, and assumes you agree to all of the bundled programs being installed. It’s always best to opt for the “custom” installation mode, which can give you a better idea of what is actually being installed, and can let you control certain aspects of the installation.

Either way, it’s best to start with the assumption that if a software or service online is “free,” that there is likely some component involved that allows the provider of that service to monetize your activity. As KrebsOnSecurity noted at the conclusion of last week’s story on a China-based proxy service called 911, the rule of thumb for transacting online is that if you’re not the paying customer, then you and/or your devices are probably the product that’s being sold to others.

Further reading on proxy services:

July 18, 2022: A Deep Dive Into the Residential Proxy Service ‘911’
June 28, 2022: The Link Between AWM Proxy & the Glupteba Botnet
June 22, 2022: Meet the Administrators of the RSOCKS Proxy Botnet
Sept. 1, 2021: 15-Year-Old Malware Proxy Network VIP72 Goes Dark
Aug. 19, 2019: The Rise of “Bulletproof” Residential Networks

A Deep Dive Into the Residential Proxy Service ‘911’

For the past seven years, an online service known as 911 has sold access to hundreds of thousands of Microsoft Windows computers daily, allowing customers to route their Internet traffic through PCs in virtually any country or city around the globe — but predominantly in the United States. 911 says its network is made up entirely of users who voluntarily install its “free VPN” software. But new research shows the proxy service has a long history of purchasing installations via shady “pay-per-install” affiliate marketing schemes, some of which 911 operated on its own.

911[.]re is one of the original “residential proxy” networks, which allow someone to rent a residential IP address to use as a relay for his/her Internet communications, providing anonymity and the advantage of being perceived as a residential user surfing the web.

From a website’s perspective, the IP traffic of a residential proxy network user appears to originate from the rented residential IP address, not from the proxy service customer. These services can be used in a legitimate manner for several business purposes — such as price comparisons or sales intelligence — but they are massively abused for hiding cybercrime activity because they can make it difficult to trace malicious traffic to its original source.

Residential proxy services are often marketed to people seeking the ability to evade country-specific blocking by the major movie and media streaming providers. But some of them — like 911 — build their networks in part by offering “free VPN” or “free proxy” services that are powered by software which turns the user’s PC into a traffic relay for other users. In this scenario, users indeed get to use a free VPN service, but they are often unaware that doing so will turn their computer into a proxy that lets others use their Internet address to transact online.

Researchers at the University of Sherbrooke in Canada recently published an analysis of 911, and found there were roughly 120,000 PCs for rent via the service, with the largest number of them located in the United States.

“The 911[.]re network uses at least two free VPN services to lure its users to install a malware-like software that achieves persistence on the user’s computer,” the researchers wrote. “During the research we identified two free VPN services that [use] a subterfuge to lure users to install software that looks legitimate but makes them part of the network. These two software are currently unknown to most if not all antivirus companies.”

The researchers concluded that 911 is supported by a “mid scale botnet-like infrastructure that operates in several networks, such as corporate, government and critical infrastructure.” The Canadian team said they found many of the 911 nodes available for rent were situated within several major US-based universities and colleges, critical infrastructures such as clean water, defense contractors, law enforcement and government networks.

Highlighting the risk that 911 nodes could pose to internal corporate networks, they observed that “the infection of a node enables the 911.re user to access shared resources on the network such as local intranet portals or other services.”

“It also enables the end user to probe the LAN network of the infected node,” the paper continues. “Using the internal router, it would be possible to poison the DNS cache of the LAN router of the infected node, enabling further attacks.”

THE INTERNET NEVER FORGETS

A review of the clues left behind by 911’s early days on the Internet paint a more complete picture of this long-running proxy network. The domain names used by 911 over the years have a few common elements in their original WHOIS registration records, including the address ustraffic@qq.com and a Yunhe Wang from Beijing.

That ustraffic email is tied to a small number of interesting domains, including browsingguard[.]com, cleantraffic[.]net, execlean[.]net, proxygate[.]net, and flashupdate[.]net.

A cached copy of flashupdate[.]net available at the Wayback Machine shows that in 2016 this domain was used for the “ExE Bucks” affiliate program, a pay-per-install business which catered to people already running large collections of hacked computers or compromised websites. Affiliates were paid a set amount for each installation of the software, with higher commissions for installs in more desirable nations, particularly Europe, Canada and the United States.

“We load only one software — it’s a Socks5 proxy program,” read the message to ExE Bucks affiliates. The website said affiliates were free to spread the proxy software by any means available (i.e. “all promotion methods allowed”). The website’s copyright suggests the ExE Bucks affiliate program dates back to 2012.

Another domain tied to the ustraffic@qq.com email in 2016 was ExeClean[.]net, a service that advertised to cybercriminals seeking to obfuscate their malicious software so that it goes undetected by all or at least most of the major antivirus products on the market.

“Our technology ensures the maximum security from reverse engineering and antivirus detections,” ExEClean promised.

Yet another domain connected to the ustraffic email is p2pshare[.]net, which advertised “free unlimited internet file-sharing platform” for those who agreed to install their software.

Still more domains associated with ustraffic@qq.com suggest 911’s proxy has been disguised as security updates for video player plugins, including flashplayerupdate[.]xyz, mediaplayerupdate[.]xyz, and videoplayerupdate[.]xyz.

The earliest version of the 911 website available from the Wayback Machine is from 2016. A sister service called proxygate[.]net launched roughly a year prior to 911 as a “free” public test of the budding new residential proxy service. “Basically using clients to route for everyone,” was how Proxygate described itself in 2016.

For more than a year after its founding, the 911 website was written entirely in Simplified Chinese. The service has only ever accepted payment via virtual currencies such as Bitcoin and Monero, as well as Alipay and China UnionPay, both payment platforms based in China.

Initially, the terms and conditions of 911’s “End User License Agreement (EULA) named a company called Wugaa Enterprises LLC, which was registered in California in 2016. Records from the California Secretary of State office show that in November 2016, Wugaa Enterprises said it was in the Internet advertising business, and had named as its CEO as one Nicolae Aurelian Mazgarean of Brasov, Romania.

A search of European VAT numbers shows the same Brasov, RO address tied to an enterprise called PPC Leads SRL (in the context of affiliate-based marketing, “PPC” generally refers to the term “pay-per-click”).

911’s EULA would later change its company name and address in 2017, to International Media Ltd. in the British Virgin Islands. That is the same information currently displayed on the 911 website.

The EULA attached to 911 software downloaded from browsingguard[.]com (tied to the same ustraffic@qq email that registered 911) references a company called Gold Click Limited. According to the UK Companies House, Gold Click Limited was registered in 2016 to a 34-year-old Yunhe Wang from Beijing City. Many of the WHOIS records for the above mentioned domains also include the name Yunhe Wang, or some variation thereof.

In a response to questions from KrebsOnSecurity, 911 said the researchers were wrong, and that 911 has nothing to do with any of the other domains mentioned above.

“We have 911 SDK link and how it works described clearly in the “Terms of use” of affiliated partners products, and we have details of how the community powered network works on our webpages,” read an email response.

“Besides that, for protecting the end users, we banned many domains’ access and blocked the vulnerable ports, e.g. spamming emails, and torrent is not possible from the 911 network,” the reply continued. “Same as scanning and many others…Accessing to the Lan network and router is also blocked. We are monitoring 911 user’s account closely, once any abnormal behavior detected, we suspend the user’s account right away.”

FORUM ACTIVITY?

911 has remained one of the most popular services among denizens of the cybercrime underground for years, becoming almost shorthand for connecting to that “last mile” of cybercrime. Namely, the ability to route one’s malicious traffic through a computer that is geographically close to the consumer whose credit card they’re about to charge at some website, or whose bank account they’re about to empty.

Given the frequency with which 911 has been praised by cybercrooks on the top forums, it was odd to find the proprietors of 911 do not appear to have created any official support account for the service on any of several dozen forums reviewed by this author going back a decade. However there are two cybercriminal identities on the forums that have responded to individual 911 help requests, and who promoted the sale of 911 accounts via their handles.

Both of these identities were active on the crime forum fl.l33t[.]su between 2016 and 2019. The user “Transfer” advertised and sold access to 911 from 2016 to 2018, amid many sales threads where they advertised expensive electronics and other consumer goods that were bought online with stolen credit cards.

In a 2017 discussion on fl.l33t[.]su, the user who picked the handle “527865713” could be seen answering private messages in response to help inquiries seeking someone at 911. That identity is tied to an individual who for years advertised the ability to receive and relay large wire transfers from China.

One ad from this user in 2016 offered a “China wire service” focusing on Western Union payments, where “all transfers are accepted in China.” The service charged 20 percent of all “scam wires,” unauthorized wire transfers resulting from bank account takeovers or scams like CEO impersonation schemes.

911 TODAY

In August 2021, 911’s biggest competitor — a 15-year-old proxy network built on malware-compromised PCs called VIP72 — abruptly closed up shop. Almost overnight, an overwhelming number of former VIP72 customers began shifting their proxy activities to 911.

That’s according to Riley Kilmer, co-founder of Spur.us — a security company that monitors anonymity services. Kilmer said 911 also gained an influx of new customers after the Jan. 2022 closure of LuxSocks, another malware-based proxy network.

“911’s user base skyrocketed after VIP72 and then LuxSocks went away,” Kilmer said. “And it’s not hard to see why. 911 and VIP72 are both Windows-based apps that operate in a similar way, where you buy private access to IPs.”

Kilmer said 911 is interesting because it appears to be based in China, while nearly all of the other major proxy networks are Russian-backed or Russian-based.

“They have two basic methods to get new IPs,” Kilmer said. “The free VPN apps, and the other is trojanized torrents. They’ll re-upload Photoshop and stuff like that so that it’s backdoored with the 911 proxy. They claim the proxy is bundled with legitimate software and that users all agree to their Terms of Service, meanwhile they can hide behind the claim that it was some affiliate who installed the software, not them.”

Kilmer said at last count, 911 had nearly 200,000 proxy nodes for sale, spanning more than 200 countries: The largest geographic concentration is the United States, where more than 42,000 proxies are currently for rent by the service.

PARTING THOUGHTS

Beware of “free” or super low-cost VPN services. Proper VPN services are not cheap to operate, so the revenue for the service has to come from somewhere. And there are countless “free” VPN services that are anything but, as we’ve seen with 911.

In general, the rule of thumb for transacting online is that if you’re not the paying customer, then you and/or your devices are probably the product that’s being sold to others. Many free VPN services will enlist users as VPN nodes for others to use, and some even offset costs by collecting and reselling data from their users.

All VPN providers claim to prioritize the privacy of their users, but many then go on to collect and store all manner of personal and financial data from those customers. Others are fairly opaque about their data collection and retention policies.

I’ve largely avoided wading into the fray about which VPN services are best, but there are so many shady and just plain bad ones out there that I’d be remiss if I didn’t mention one VPN provider whose business practices and transparency of operation consistently distinguish them from the rest. If maintaining your privacy and anonymity are primary concerns for you as a VPN user, check out Mullvad.net.

Let me make clear that KrebsOnSecurity does not have any financial or business ties to this company (for the avoidance of doubt, this post doesn’t even link to them). I mention it only because I’ve long been impressed with their candor and openness, and because Mullvad goes out of its way to discourage customers from sharing personal or financial data.

To that end, Mullvad will even accept mailed payments of cash to fund accounts, quite a rarity these days. More importantly, the service doesn’t ask users to share phone numbers, email addresses or any other personal information. Nor does it require customers to create passwords: Each subscription can be activated just by entering a Mullvad account number (woe to those who lose their account number).

I wish more companies would observe this remarkably economical security practice, which boils down to the mantra, “You don’t have to protect what you don’t collect.”

Update, July 24, 11:15 a.m. ET: 911’s homepage now includes a banner saying the service has halted new registrations and payments. “We are reviewing our network and adding a series of security measures to prevent misuse of our services,” the message reads. “Proxy balance top-up and new user registration are closed. We are reviewing every existing user, to ensure their usage is legit and [in] compliance with our Terms of Service.”

Update, July 30, 10:07 a.m. ET: 911 announced on July 28 that it is permanently closing down, following a series of data breaches this month that 911 says resulted in the deletion of customer data.

The Link Between AWM Proxy & the Glupteba Botnet

On December 7, 2021, Google announced it was suing two Russian men allegedly responsible for operating the Glupteba botnet, a global malware menace that has infected millions of computers over the past decade. That same day, AWM Proxy — a 14-year-old anonymity service that rents hacked PCs to cybercriminals — suddenly went offline. Security experts had long seen a link between Glupteba and AWM Proxy, but new research shows AWM Proxy’s founder is one of the men being sued by Google.

Launched in March 2008, AWM Proxy quickly became the largest service for crooks seeking to route their malicious Web traffic through compromised devices. In 2011, researchers at Kaspersky Lab showed that virtually all of the hacked systems for rent at AWM Proxy had been compromised by TDSS (a.k.a TDL-4 and Alureon), a stealthy “rootkit” that installs deep within infected PCs and loads even before the underlying Windows operating system boots up.

In March 2011, security researchers at ESET found TDSS was being used to deploy Glupteba, another rootkit that steals passwords and other access credentials, disables security software, and tries to compromise other devices on the victim’s network — such as Internet routers and media storage servers — for use in relaying spam or other malicious traffic.

Like its predecessor TDSS, Glupteba is primarily distributed through “pay-per-install” or PPI networks, and via traffic purchased from traffic distribution systems (TDS). Pay-per-install networks try to match cybercriminals who already have access to large numbers of hacked PCs with other crooks seeking broader distribution of their malware.

In a typical PPI network, clients will submit their malware—a spambot or password-stealing Trojan, for example —to the service, which in turn charges per thousand successful installations, with the price depending on the requested geographic location of the desired victims. One of the most common ways PPI affiliates generate revenue is by secretly bundling the PPI network’s installer with pirated software titles that are widely available for download via the web or from file-sharing networks.

Over the past decade, both Glupteba and AWM Proxy have grown substantially. When KrebsOnSecurity first covered AWM Proxy in 2011, the service was selling access to roughly 24,000 infected PCs scattered across dozens of countries. Ten years later, AWM Proxy was offering 10 times that number of hacked systems on any given day, and Glupteba had grown to more than one million infected devices worldwide.

There is also ample evidence to suggest that Glupteba may have spawned Meris, a massive botnet of hacked Internet of Things (IoT) devices that surfaced in September 2021 and was responsible for some of the largest and most disruptive distributed denial-of-service (DDoS) attacks the Internet has ever seen.

But on Dec. 7, 2021, Google announced it had taken technical measures to dismantle the Glupteba botnet, and filed a civil lawsuit (PDF) against two Russian men thought to be responsible for operating the vast crime machine. AWM Proxy’s online storefront disappeared that same day.

AWM Proxy quickly alerted its customers that the service had moved to a new domain, with all customer balances, passwords and purchase histories seamlessly ported over to the new home. However, subsequent takedowns targeting AWM Proxy’s domains and other infrastructure have conspired to keep the service on the ropes and frequently switching domains ever since.

Earlier this month, the United States, Germany, the Netherlands and the U.K. dismantled the “RSOCKS” botnet, a competing proxy service that had been in operation since 2014. KrebsOnSecurity has identified the owner of RSOCKS as a 35-year-old from Omsk, Russia who runs the world’s largest forum catering to spammers.

Shortly after last week’s story on the RSOCKS founder, I heard from Riley Kilmer, co-founder of Spur.us, a startup that tracks criminal proxy services. Kilmer said RSOCKS was similarly disabled after Google’s combined legal sneak attack and technical takedown targeting Glupteba.

“The RSOCKS website gave you the estimated number of proxies in each of their subscription packages, and that number went down to zero on Dec. 7,” Kilmer said. “It’s not clear if that means the services were operated by the same people, or if they were just using the same sources (i.e., PPI programs) to generate new installations of their malware.”

Kilmer said each time his company tried to determine how many systems RSOCKS had for sale, they found each Internet address being sold by RSOCKS was also present in AWM Proxy’s network. In addition, Kilmer said, the application programming interfaces (APIs) used by both services to keep track of infected systems were virtually identical, once again suggesting strong collaboration.

“One hundred percent of the IPs we got back from RSOCKS we’d already identified in AWM,” Kilmer said. “And the IP port combinations they give you when you access an individual IP were the same as from AWM.”

In 2011, KrebsOnSecurity published an investigation that identified one of the founders of AWM Proxy, but Kilmer’s revelation prompted me to take a fresh look at the origins of this sprawling cybercriminal enterprise to determine if there were additional clues showing more concrete links between RSOCKS, AWM Proxy and Glupteba.

IF YOUR PLAN IS TO RIP OFF GOOGLE…

Supporting Kilmer’s theory that AWM Proxy and RSOCKS may simply be using the same PPI networks to spread, further research shows the RSOCKS owner also had an ownership stake in AD1[.]ru, an extremely popular Russian-language pay-per-install network that has been in operation for at least a decade.

Google took aim at Glupteba in part because its owners were using the botnet to divert and steal vast sums in online advertising revenue. So it’s more than a little ironic that the critical piece of evidence linking all of these operations begins with a Google Analytics code included in the HTML code for the original AWM Proxy back in 2008 (UA-3816536).

That analytics code also was present on a handful of other sites over the years, including the now-defunct Russian domain name registrar Domenadom[.]ru, and the website web-site[.]ru, which curiously was a Russian company operating a global real estate appraisal business called American Appraisal.

Two other domains connected to that Google Analytics code — Russian plastics manufacturers techplast[.]ru and tekhplast.ru — also shared a different Google Analytics code (UA-1838317) with web-site[.]ru and with the domain “starovikov[.]ru.”

The name on the WHOIS registration records for the plastics domains is an “Alexander I. Ukraincki,” whose personal information also is included in the domains tpos[.]ru and alphadisplay[.]ru, both apparently manufacturers of point-of-sale payment terminals in Russia.

Constella Intelligence, a security firm that indexes passwords and other personal information exposed in past data breaches, revealed dozens of variations on email addresses used by Alexander I. Ukraincki over the years. Most of those email addresses start with some variation of “uai@” followed by a domain from one of the many Russian email providers (e.g., yandex.ru, mail.ru). [Full disclosure: Constella is currently an advertiser on this website].

But Constella also shows those different email addresses all relied on a handful of passwords — most commonly “2222den” and “2222DEN.” Both of those passwords have been used almost exclusively in the past decade by the person who registered more than a dozen email addresses with the username “dennstr.”

The dennstr identity leads to several variations on the same name — Denis Strelinikov, or Denis Stranatka, from Ukraine, but those clues ultimately led nowhere promising. And maybe that was the point.

Things began looking brighter after I ran a search in DomainTools for web-site[.]ru’s original WHOIS records, which shows it was assigned in 2005 to a “private person” who used the email address lycefer@gmail.com. A search in Constella on that email address says it was used to register nearly two dozen domains, including starovikov.ru and starovikov[.]com.

A cached copy of the contact page for Starovikov[.]com shows that in 2008 it displayed the personal information for a Dmitry Starovikov, who listed his Skype username as “lycefer.”

Finally, Russian incorporation documents show the company LLC Website (web-site[.]ru)was registered in 2005 to two men, one of whom was named Dmitry Sergeevich Starovikov.

Bringing this full circle, Google says Starovikov is one of the two operators of the Glupteba botnet:

Mr. Starovikov did not respond to requests for comment. But attorneys for Starovikov and his co-defendant last month filed a response to Google’s complaint in the Southern District of New York, denying (PDF) their clients had any knowledge of the scheme.

Despite all of the disruption caused by Google’s legal and technical meddling, AWM is still around and nearly as healthy as ever, although the service has been branded with a new name and there are dubious claims of new owners. Advertising customer plans ranging from $50 a day to nearly $700 for “VIP access,” AWM Proxy says its malware has been running on approximately 175,000 systems worldwide over the last 24 hours, and that roughly 65,000 of these systems are currently online.

Meanwhile, the administrators of RSOCKS recently alerted customers that the service and any unspent balances will soon be migrated over to a new location.

Many people seem to equate spending time, money and effort to investigate and prosecute cybercriminals with the largely failed war on drugs, meaning there is an endless supply of up-and-coming crooks who will always fill in any gaps in the workforce whenever cybercriminals face justice.

While that may be true for many low-level cyber thieves today, investigations like these show once again how small the cybercriminal underground really is. It also shows how it makes a great deal of sense to focus efforts on targeting and disrupting the relatively small number of established hackers who remain the real force multipliers of cybercrime.