Showing posts with label Microsoft. Show all posts
Showing posts with label Microsoft. Show all posts

Wednesday, March 10, 2021

Microsoft Exchange: Patching Too Late If Already Compromised

On March 2, 2021, Microsoft accused a Chinese APT group which they name Hafnium of compromising 30,000 Exchange servers.  They announced four security vulnerabilities, known as 0-days, which refers to the fact that attackers had a reliable means of exploiting the vulnerability for which there was no patch.  In case your organization didn't go into full panic mode, GO PULL THE FIRE ALARM!  THIS IS SERIOUS!

Tom Burt, Microsoft's VP of Customer Security & Trust, released a blog post about Hafnium: New Nation-State Cyberattacks. Microsoft describes Hafnium as "primarily targeting entities in the United States for the purpose of exfiltrating information from a number of industry sectors, including infectious disease researchers, law firms, higher education institutions, defense contractors, policy think tanks and NGOs." According to my favorite APT Cross-reference chart, maintained by Florian Roth (Twitter: @Cyb3rOps) Hafnium is also referred to by Symantec as Ant. (They chose the name because one of the common webshells used for post exploitation was regularly hit by a web browser using the user agent "antSword/v2.1".  Both AntSword (中国蚁剑 ) and ChinaChopper (中国菜刀) are popular webshells used by Chinese attackers for many years.  

FireEye says there is no reason to believe the activity is limited to one threat actor and refers to the clusters of attacks as UNC2639, UNC2640, UNC2643.

FireEye associates UNC2639 with activity from IP addresses 165.232.154.116 and 182.18.152.105, both active at the time of the Microsoft announcement (March 2 and March 3).

FireEye associates UNC2640 with activity involving "web shell" files named "help.aspx" (MD5 4b3039cf227c611c45d2242d1228a121) and "iisstart.aspx" (MD5 0fd9bffa49c76ee12e51e3b8ae0609ac)

FireEye associates UNC2643 with the deployment of a Cobalt Strike Beacon (MD5 79eb217578bed4c250803bd573b10151) and the IP addresses 89.34.111.11 and 86.105.18.116.

FireEye says they began seeing this activity in January, which matches the reports from Microsoft that they were notified of this activity by security firm Volexity in January, however DEVCORE Research Team gets credit for trying to exploit marketing of the bug by calling the attack "ProxyLogon" and making a sexy webpage and logo for the attack, a la HeartBleed.  Fortunately, that really hasn't caught on, however their timeline is still very interesting. They found the first bug 10DEC2020 and the second 30DEC2020 and reported both to Microsoft on 05JAN2021 (as Tweeted by their Taiwanese researcher, Orange Tsai.)

Symantec makes clear that the actor which they call Ant (and Microsoft calls Hafnium) is definitely no longer the only attacker using these vulnerabilities. Symantec's diagram of the attack is useful:

Symantec's attack flow diagram 

Brian Krebs interviewed several researchers about the attack, including Steven Adair, who says his company, Volexity, has been seeing the bug since 06JAN2021.  See Krebs on Security: "At Least 30,000 US Organizations Newly Hacked via Holes in Microsoft's Email Software."

As Krebs and others have since pointed out, while 30,000 US-based organizations were known to be victims of Hafnium/Ant, now that the vulnerability is known, the attacks have grown to an astronomical number.  Why is this a problem?  The companies most likely to be running their own unpatched mail servers are also the least likely to be clueful enough to patch.

Both Forbes and WIRED now say that hundreds of thousands of servers have been compromised and the compromise count at one point was growing by "thousands per hour."


What To Do?  PATCH! (But it is quite possibly too late...)

Obviously, the most important thing to do is apply Microsoft's patches.  However it is VERY IMPORTANT TO UNDERSTAND that you may already be compromised.  Patching DOES NOT make you "un-hacked!"  Patch, but also follow the guidance from CISA on determining if you are already hacked.

The vulnerabilities are listed here, each linking to the Microsoft security alert associated with the CVE.


Unfortunately, smaller organizations tend not to patch, and rogue organizations within large organizations often run their own Exchange servers rather than following guidance to centralize. In a presentation I did for the Merchant Risk Council back in September 2020, we talked about the fact that CISA had put out a critical alert related to Office 365, calling it a "Top 10 Routinely Exploited Vulnerability" as well as its own alert, CISA Alert AA20-120A.  In that talk, we also mentioned how Rapid7's Tom Sellers had warned about unpatched Exchange Servers.  Sellers was actually talking about the "Critical" Exchange Server bug CVE-202-0688.  

In Rapid7's look at the data, "Phishing for SYSTEM on Microsoft Exchange (CVE 2020-0688)" originally published on 06APR2020 explained that a 24MAR2020 scan of the Internet found 357,629 vulnerable servers, 82.5% of those reachable from the public Internet, were unpatched for a CRITICAL vulnerability with a patch available since 11FEB2020.  EIGHT MONTHS LATER, Rapid7 repeated the test, and still found that 61% of those servers were still online and still vulnerable!  Further, 31,000 servers had not been patched since 2012, and 800 servers had NEVER been patched!

What do you think the chances are that they suddenly became patch-conscious on 02MAR2021?

It is quite likely, in this author's opinion, that MOST Internet-facing Exchange servers have been compromised.  How do you test to see if you are one of them?  Read on ...

WHAT TO DO?  SEE IF YOU ARE HACKED!



The CyberSecurity & Infrastructure Security Agency, CISA, part of the Department of Homeland Security, has provided comprehensive information on how to detect the attack, including a nice guide on how to use FTK Imager to capture memory from your Exchange Server and where to look for evidence of being compromised.

Please thoroughly review their recommendations found as Alert AA21-062A.

Many of their indicators come from Volexity, who also shares a video explaining the attack in their blog post from 02MAR2021, "Operation Exchange Marauder."  It should be noted that neither of the IP addresses from FireEye are included on this list.

In addition to the CISA guidance, Microsoft has released a script which can be run on your Exchange Server to look for signs of being compromised.  Their script is described in their Hafnium Targeting Exchange Servers blog post, but a direct link to the script is: 

https://github.com/microsoft/CSS-Exchange/tree/main/Security

This script scans the HttpProxy logs, the Exchange logs, and the Windows Application event logs for signs of exploitation.  Hopefully the bad guys haven't WIPED the logs!






Saturday, December 22, 2018

126 Arrests: The Emergence of India's Cyber Crime Detectives Fighting Call Center Scams

The Times of India reports that police have raided a call center in Noida Sector 63 where hundreds of fraud calls were placed every day to Americans and Canadians resulting in the theft of $50,000 per day.

 The scammers had rented four floors of a building being operated by two scammers from Gurgaon, Narendra Pahuja and Jimmy Ashija. Their boss, who was not named by the police, allegedly operates at least five call centers. In the raid this week, 126 employees were arrested and police seized 312 workstations, as well as Rs 20 lakh in cash (about $28,500 USD).

Times of India photo 


Noida police have been cooperating very well with international authorities, as well as Microsoft, leading to more than 200 people arrested in Noida and "scores" of fake call centers shut down, including four in Sector 63.  (In a case just last month, another call center was said to have stolen from 300 victims, after using online job sites Shine.com and VintechJobs.com to recruit young money seekers by having them work conducting the scams. )

In the current scam, callers already had possession of the victim's Social Security Number and full name.  This information was used to add authority to their request, which got really shady really fast.  The victim was instructed to purchase Apple iTunes Gift Cards, or Google Play Gift Cards, scratch the numbers, and read them to the call center employee.  The money was laundered through a variety of businesses in China and India before cashing out to bank accounts belonging to Pahuja and Ashija.

 Go to Tweet
Noida police are advancing in their Cyber Crime skills!

As more and more cyber crime enterprises spring up in India, the assistance of their new Centers for Cyber Crime Investigation thtat are becoming more critical to stopping fraud against Americans:

We applaud the Center for Cyber Crime Investigation in Noida


The US Embassy was quick to acknowledge the support of the newest cyber crime partners of the United States after their action at the end of November:

US Embassy to India thanks the Noida and Gurgaon Police for their help!
Another recent Times of India story from November 30, 2018, "Bogus Call Centres and Pop-up Virus Alerts - a Global Cyber Con Spun up in NCR" [NCR = National Capital Region] had more details of this trend, including this graphic:


That's at least 50 call centers shutdown just in these two regions, but with this weeks' 126 arrests being the culmination of an on-going investigation, receiving data from both the FBI and Microsoft.

Local news of India reported the names of some of the gang members held in the November 29-30th action in their story नोएडा: बड़ी कंपनियों में नौकरी दिलाने के नाम पर करते थे धोखाधड़ी, 8 गिरफ्तार (Noida: Fraud, 8 arrested for giving fake jobs in the name of big companies).

Sontosh Gupta, who was the ring leader, was previously employed by an online job site, but then created his own site,  vintechjobs (dot) com, which he used to attract call center employees, many of whom were duped into serving as his scammer army without ever being compensated for their work!

Others arrested then included Mohan Kumar, Paritosh Kumar, Jitendra Kumar, Victor, Himanshu, Ashish Jawla, and Jaswinder.

During that same two day raid, police swept through at least sixteen other call centers, according to this New York Times story, "That Virus Alert on Your Computer? Scammers in India May Be Behind It"
Ajay Pal Sharma, the senior superintendent of police, told the NYT that 50 of his officers swept through eight different call centers in Gautam Budh Nagar as part of the case.  Microsoft's Digital Crimes Unit told the Times that with 1.2 million people generating $28 Billion in India working for call centers, it isn't hard to disguise the shady callers among the legitimate businesses.

The problem is not unique to Delhi and the National Capital Region suburbs that are the current focus.  Back in July, Mumbai was in the headlines, as a massive IRS-imitating Call Center ring was broken up with the help of more great cyber crime investigators from India:

Madan Ballal, Thane Crime Branch, outside Mumbai
Police Inspector Madan Ballal had his story told as the focus of an article in Narratively, "This Indian Cop Took Down a Massive IRS Call-Center Scam".

Much more investigating and arresting needs to be done, but it is a great sign that the problem is now receiving help from an emerging new generation of Indian Cybercrime Detectives!