Connect with us

Press Release

Samsung Galaxy S22 hacked in 55 seconds

Published

on

Samsung Galaxy S22 hacked in 55 seconds

​Comcast Xfinity customers report their accounts being hacked in widespread attacks that bypass two-factor authentication. These compromised accounts are then used to reset passwords for other services, such as the Coinbase and Gemini crypto exchanges.

Starting on December 19th, many Xfinity email users began receiving notifications that their account information had been changed. However, when attempting to access the accounts, they could not log in as the passwords had been changed.

After regaining access to the accounts, they discovered they had been hacked and a secondary email at the disposable @yopmail.com domain was added to their profile.

Similar to Gmail, Xfinity allows customers to configure a secondary email address to be used for account notifications and password resets in the event they lose access to their Xfinity account.

BleepingComputer first learned of these account hacks after numerous Xfinity customers reached out to us to share their experiences. In addition, other customers shared similar reports on Reddit [1, 2], Twitter [1, 2, 3], and Xfinity’s own support forum.

All Xfinity customers we spoke to said they have two-factor authentication enabled on their accounts, yet the threat actors could bypass it and log in to their accounts.

“Someone was able to reset my password and change personal account information, they bypassed 2FA. the email they setup was xxxxxxxx@yopmail.com,” explained an Xfinity customer on Reddit.

2FA bypass allegedly circulating privately

A researcher has told BleepingComputer that the attacks are being conducted through credential stuffing attacks to determine the login credentials for Xfinity attacks.

Once they gain access to the account and are prompted to enter their 2FA code, the attackers allegedly use a privately circulated OTP bypass for the Xfinity site that allows them to forge successful 2FA verification requests.

Once logged into the account, they can change the secondary email to the @yopmail.com account and perform password resets.

The main Xfinity email will also receive a notification that their information was changed, but as the password has been changed as well, will be unable to access it.

Once they gain full access to an Xfinity email account, the threat actors attempt to breach further online services used by the customer, verifying password reset requests to the now compromised email account.

BleepingComputer has been told by some of the affected customers that the hackers attempted to reset passwords at DropBox, Evernote, and the Coinbase and Gemini cryptocurrency exchanges.

While BleepingComputer has been unable to verify the legitimacy of this OTP bypass independently and whether it has been used in the reported hacks, it would explain how the threat actors can gain access to accounts with 2FA enabled.

BleepingComputer reached out to Comcast press contacts several times this week but has yet to receive a reply to our emails.

However, an Xfinity customer posted on Reddit that the company is aware of the account breaches and looking for the source of the hacks.

“I spoke to a second person in the xfinity security department that told me not to worry about the fraudulent yopmail account on my xfinity account and indicated that this had happened with many (maybe all) xfinity accounts,” a user posted to Reddit about the hacks.

“She indicated that xfinity is still working to find the source of the hack. Apparently this this is a much more widespread issue than is being reported. It does not seem that xfinity e-mail is secure at this time.”

 

Press Release

Microsoft fumbles supply chain and acknowledges signing rootkit malware.

Published

on

Microsoft fumbles supply chain and acknowledges signing rootkit malware.

As of right now, Microsoft has admitted to signing a malicious driver that is disseminated in gaming contexts.

This “Netfilter”-named driver is actually a rootkit that has been seen interacting with Chinese C2 IP addresses.

Last week, the whole infosec. community joined G Data malware specialist Karsten Hahn in tracking down and analysing the malicious drivers that bore the Microsoft logo.

This incident exposed vulnerabilities to software supply-chain security once more, but this time it was caused by a flaw in the code-signing procedure used by Microsoft.

Rootkit “Netfilter” driver is Microsoft-signed.
A Microsoft signed driver dubbed “Netfilter” was detected last week by G Data’s cybersecurity alert systems as what at first glance appeared to be a false positive, but wasn’t.

The driver in question was observed interacting with C&C IPs based in China, which had no valid functionality and raised red flags.

This is when Karsten Hahn, a malware analyst at G Data, disclosed this publicly and contacted Microsoft at the same time:

Since Windows Vista, all code that operates in kernel mode must be tested and certified before being made available to the public in order to maintain the stability of the operating system.

According to Hahn, “Drivers without a Microsoft certificate cannot be deployed by default.”

At that time, BleepingComputer started tracking C2 URL behaviour and approached Microsoft for a comment.

A list of further routes (URLs), denoted by the pipe (“|”) symbol, are returned by the first C2 URL:

Each of these, in Hahn’s opinion, has a function:

The URL that ends in “/p” refers to proxy settings, “/s” offers encoded redirection IPs, “/h?” is for getting CPU-ID, “/c” offered a root certificate, and “/v?” refers to the malware’s self-updating capabilities.
For instance, as observed by BleepingComputer, the malicious Netfilter driver in question (residing at “/d3”) was accessible via the “/v?” path at the following URL:

After thoroughly examining the driver, the G Data researcher came to the conclusion that it was malware.

In a thorough blog post, the researcher examined the driver, its ability to self-update, and Indicators of Compromise (IOCs).

According to Hahn, the sample features a self-update routine that transmits its own MD5 hash to the server via the URL hxxp:/110.42.4.180:2081/v?v=6&m=.

An illustration of a request would be as follows:

hxxp:/110.42.4.180:2081/v?v=6&m=921fa8a5442e9bf3fe727e770cded4ab
“The server then replies with either ‘OK’ if the sample is current or the URL for the most recent sample, such as hxxp:/110.42.4.180:2081/d6. As a result, the malware replaces its own file “further information from the researcher

Other malware specialists like Johann Aydinbas, Takahiro Haruyama, and Florian Roth worked with Hahn during his analysis.

Roth has offered YARA rules for recognising them in your network environments after being able to compile the list of samples in a spreadsheet.

Microsoft is looking at a bad actor who spreads harmful drivers inside of gaming environments.

“In order to be certified by the Windows Hardware Compatibility Program, the actor supplied drivers. A third party created the drivers.”

Microsoft stated yesterday, “We have stopped the account and checked their uploads for additional indicators of malware.”

Microsoft claims that the threat actor primarily targeted the gaming industry in China with these malicious drivers and that there is currently no evidence that enterprise environments have been impacted.

Microsoft is waiting before blaming nation-state actors for this incident.

Sophisticated threat actors may take advantage of falsely signed binaries to help launch extensive software supply-chain attacks.

A well-known event in which code-signing certificates were taken from Realtek and JMicron to assist the comprehensive Stuxnet attack on Iran’s nuclear programme.

However, this specific instance has shown flaws in a reliable code-signing procedure, which threat actors have exploited to obtain Microsoft-signed code without jeopardising any certifications.

Continue Reading

Press Release

FlexBooker reports a data breach, affecting more than 3.7 million accounts.

Published

on

FlexBooker reports a data breach, affecting more than 3.7 million accounts.

In an attack just before the holidays, the accounts of over three million customers of the American appointment scheduling service FlexBooker were taken, and they are now being exchanged on hacker forums.

The same hackers are also selling databases they claim to be from two other organisations: the Australian case management system rediCASE and the racing media outlet Racing.com.

Holiday breaches before
A few days before Christmas, there were supposedly three breaches, and the intruder posted the information on a hacking forum.

A popular programme for booking appointments and syncing employee calendars, FlexBooker, appears to be the source of the most recent data dump.

Owners of any company that needs to plan appointments, such as accountants, barbers, doctors, mechanics, lawyers, dentists, gyms, salons, therapists, trainers, spas, and the list goes on, are among FlexBooker’s clients.

The group claiming responsibility for the attack appears to go by the name of Uawrongteam, and they published links to files and archives containing personal information, including pictures, driver’s licences, and other IDs.

The database, according to Uawrongteam, has a table with 10 million lines of client data, including everything from payment forms and charges to pictures taken for driver’s licences.

Names, emails, phone numbers, password salt, and hashed passwords are among the database’s “juicy columns,” according to the actor.

Customers of FlexBooker have received a data breach notification that confirms the attack and that data on the service’s Amazon cloud storage system was “accessed and downloaded” by the intruders.

The letter states that “our account on Amazon’s AWS servers was compromised on December 23, 2021, starting at 4:05 PM EST,” adding that the attackers did not obtain “any credit card or other payment card information.”

FlexBooker advised consumers to be on the lookout for strange or fraudulent activities, and to monitor account statements and credit reports.

For further information, the developer also directed users to a report on a distributed denial-of-service (DDoS) attack. It was then determined that some customers’ personal information had been obtained by the hackers.

The FlexBooker assault exposed email addresses, names, partial credit card information, passwords, and phone numbers for more than 3.7 million users, according to the data breach reporting service Have I Been Pwned.

Prior to FlexBooker, the threat actor known as Uawrongteam distributed links to material that was purportedly taken from Racing.com, a digital television station that broadcasts horse racing and offers news, stats, and event calendars associated with the sport.

The data from the Redbourne Gang’s rediCASE Case Management Software, which is utilised by numerous enterprises in addition to health and community agencies, looks to be another target of the same group.

Continue Reading

Press Release

Rapyd, a “fintech-as-a-service” provider, to acquire Iceland-based Valitor, which establishes in-store and on the internet payments technologies, for $100M (Omar Faridi/Crowdfund Expert).

Published

on

acquire Iceland-based Valitor

Rapyd, a “fintech-as-a-service” provider, to acquire Iceland-based Valitor, which develops in-store and online payments technologies, for $100M (Omar Faridi/Crowdfund Insider)

Omar Faridi / Crowdfund Insider:
Rapyd, a “fintech-as-a-service” provider, to acquire Iceland-based Valitor, which develops in-store and online payments technologies, for $100M  —  – Twitter- Facebook- LinkedIn- Pinterest- Reddit- HackerNews- Telegram- Weibo- Email- Print- Subscribe

Continue Reading

Trending