Executive Summary

• This month, large scale phishing attacks against two German banking associations (Sparkasse and Volksbanken und Raiffeisenbanken)

Summary

In this installment of our monthly email threat review, we present an overview of the email-based threats observed in September 2021 and compare them to the previous month’s threats.

The report provides insights into:

• Unwanted emails by category
• File types used in attacks
• Industry Email Threat Index
• Attack techniques
• Impersonated company brands and organizations
• Ransomleaks

Unwanted emails by category

The following table shows the distribution of unwanted emails per category.

Email category	%
Rejected	80.83
Spam	14.15
Threat	4.07
AdvThreat	0.91
Content	0.03

The following time histogram shows the email volume per category per day.

The spike in rejected emails around 2021-09-23 can be attributed to a monthly reoccurring sextortion scam email campaign written in the German language that we observe each month.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain harmful content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as the sender’s identity, and the emails are not analyzed further.

File types used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicious emails)	%
HTML	37.2
Archive	28.8
PDF	11.3
Excel	6.4
Disk image files	4.6
Other	4.0
Word	3.7
Executable	3.5
Powerpoint	0.2
Script file	0.2

The following time histogram shows the email volume per file type used in attacks per 7 days.

We see the continuation of last month’s observation concerning the increase in HTML file attachments (.htm, .html, etc.) used in attacks. On closer analysis, the growth can be attributed to multiple campaigns using HTML files for phishing by having the phishing website attached directly to the email¹ (thus circumventing URL filters). We already report on this technique in our blog.

Industry Email Threat Index

The following table shows our Industry Email Threat Index calculated based on the number of threat emails compared to each industry’s clean emails received (in median).

Industries	Share of threat in threat and clean emails
Manufacturing industry	6.0
Research industry	5.7
Media industry	5.3
Healthcare industry	5.2
Automotive industry	5.1
Education industry	5.1
Transport industry	5.0
Construction industry	4.9
Mining industry	4.9
Retail industry	4.4

The following bar chart visualizes the email-based threat posed to each industry.

We observe an overall increase in threat emails per clean email received for all industry sectors.

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, we calculate the percent share of threat emails from each organization’s threat and clean emails to compare organizations. We then calculate the median of these percent values for all organizations within the same industry to form the industry’s final threat score.

Attack techniques

The following table shows the attack techniques used in attacks.

Attack technique	%
Other	39.0
Phishing	29.1
Impersonation	10.3
URL	10.0
Advance-fee scam	3.5
Extortion	3.1
Executable in archive/disk-image	3.1
Maldoc	1.9

The following time histogram shows the email volume per attack technique used per hour.

Impersonated company brands and organizations

The following table shows which company brands and organizations our systems detected most in impersonation attacks.

Impersonated brand or organization	%
Sparkasse	19.7
Volks- und Raiffeisenbank	15.0
Deutsche Post / DHL	13.5
DocuSign	12.5
Other	11.3
Amazon	8.9
PayPal	4.6
LinkedIn	1.9
Microsoft	1.7
UPS	1.3
HSBC	1.1

The following histogram shows the email volume for company brands and organizations detected in impersonation attacks per hour.

There is a significant increase in campaigns impersonating the German banks Sparkasse and Volks- und Raiffeisenbank. We detected multiple different phishing email campaigns targeting these two German banks.

One example phishing email impersonating Sparkasse:

One example phishing email impersonating Volks- und Raiffeisenbank:

In all variants of the large-scale campaigns, the banks allegedly changed their (security) processes and the user needs to log in to confirm the change to retain access to their bank account.

Ransomleaks

Threat actors continue to leak data stolen from ransomware victims to pressure them to pay for decrypting the files and not publishing sensible data. We observed the following number of leaks on ransomware leak sites:

Leak site	Number of victim data leaks
Conti	40
LockBit 2.0	34
Pysa	15
Everest	8
Vice Society	7
Hive	5
REvil	5
RansomEXX	3
Lorenz	1

The following bar chart visualizes the number of victim data leaks per leak site.

On 2021-09-16, an unnamed law enforcement agency obtained the decryption keys for the REvil ransomware and BitDefender released a decryptor.

On 2021-09-21, the Department of the Treasury’s Office of Foreign Assets Control’s (OFAC) added SUEX OTC, S.R.O. (SUEX), a virtual currency exchange, to its OFAC sanction list for its part in facilitating financial transactions for ransomware actors. Analysis of known SUEX transactions shows that over 40% of SUEX’s known transaction history is associated with illicit actors related to ransomware.²

This month LockBit 2.0 added a DDOS protection CAPTCHA to their site.

Cl0p ransomware has also added a CAPTCHA to their leak site, preventing automatic monitoring of announcements.

References

• ¹ https://www.hornetsecurity.com/en/security-informationen-en/html-phishing-asking-for-the-password-twice/
• ² https://home.treasury.gov/news/press-releases/jy0364

Executive Summary

• We observed an increase in HTML attachment-based phishing.

Summary

In this installment of our monthly email threat review, we present an overview of the email-based threats observed in August 2021 and compare them to the previous month’s threats.

The report provides insights into:

• Unwanted emails by category
• File types used in attacks
• Industry Email Threat Index
• Attack techniques
• Impersonated company brands and organizations
• Ransomleaks

Unwanted emails by category

The following table shows the distribution of unwanted emails per category.

Email category	%
Rejected	83.19
Spam	12.92
Threat	3.02
AdvThreat	0.83
Content	0.03

The following time histogram shows the email volume per category per day.

The spike in rejected emails around 2021-07-31 can be attributed to the monthly reoccurring sextortion scam email campaign written in the German language that regularly causes spikes in rejected emails.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain harmful content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as the sender’s identity, and the emails are not analyzed further.

File types used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicious emails)	%
Archive	33.7
HTML	25.1
PDF	14.9
Excel	8.7
Word	5.1
Other	4.3
Executable	4.0
Disk image files	3.8
Script file	0.3
Powerpoint	0.1
Email	0.0
LNK file	0.0

The following time histogram shows the email volume per file type used in attacks per 7 days.

We can detect an increase of HTML file attachments (.htm, .html, etc.) used in attacks. On closer analysis, the increase can be attributed to multiple campaigns using HTML files for phishing by having the phishing website attached directly to the email¹ (thus circumventing URL filters). We already report on this technique in our blog.

Industry Email Threat Index

The following table shows our Industry Email Threat Index calculated based on the number of threat emails compared to each industry’s clean emails received (in median).

Industries	Share of threat in threat and clean emails
Manufacturing industry	4.3
Entertainment industry	3.7
Research industry	3.5
Media industry	3.4
Healthcare industry	3.4
Mining industry	3.3
Transport industry	3.3
Education industry	3.0
Hospitality industry	2.9
Automotive industry	2.8

The following bar chart visualizes the email-based threat posed to each industry.

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, we calculate the percent share of threat emails from each organization’s threat and clean emails to compare organizations. We then calculate the median of these percent values for all organizations within the same industry to form the industry’s final threat score.

Attack techniques

The following table shows the attack techniques used in attacks.

Attack technique	%
Other	40.2
Phishing	27.8
URL	11.2
Impersonation	5.8
Advance-fee scam	4.8
Executable in archive/disk-image	4.0
Extortion	3.7
Maldoc	2.4
LNK	0.0

The following time histogram shows the email volume per attack technique used per hour.

The increase in malicious documents at the end of the month can be attributed to backscatter from a Formbook malspam campaign spoofing the email address of one of our customers. Many of the bounce messages contained the malicious documents of the campaign and were detected by our filters.

Impersonated company brands and organizations

The following table shows which company brands and organizations our systems detected most in impersonation attacks.

Impersonated brand or organization	%
DocuSign	18.2
Deutsche Post / DHL	17.9
Other	16.8
Amazon	12.6
PayPal	10.0
Microsoft	2.8
Volks- und Raiffeisenbank	2.6
HSBC	2.0
LinkedIn	1.8
O2	1.4
Santander	1.1
Tesco	1.1

The following histogram shows the email volume for company brands and organizations detected in impersonation attacks per hour.

It’s a constant stream of phishing and other attacks impersonating big brands and organizations to entice recipients to open the emails.

Ransomleaks

Threat actors continue to leak data stolen from ransomware victims to pressure them to pay for decrypting the files and not publishing sensible data. We observed the following number of leaks on ransomware leak sites:

Leak site	Number of victim data leaks
LockBit 2.0	87
Conti	46
Hive	27
Pysa	16
Everest	5
Cuba	4
RansomEXX	4
LV	3
Vice Society	3
Lorenz	2
Cl0p	1
RagnarLocker	1
Suncrypt	1
Xing Team	1

The following bar chart visualizes the number of victim data leaks per leak site.

On 2021-08-12, the SynAck ransomware operation renamed itself to El_Cometa. Additionally, the decryption keys for the SynAck ransomware were released. ² Our monitoring registered a total of 15 victims on SynAck’s leak site in June and Juli. The operation under the SynAck name was, therefore, rather short-lived.

According to reports³ the Ragnarok ransomware shut down its operation on 2021-08-27. Our monitoring saw the Ragnarok leak site last online on 2021-05-17. In total, the groups leak site published data of 22 victims. The operators behind the Ragnarok ransomware have also released a universal decryption key.

This practice of releasing decryption keys after shutting a ransomware operation down is widespread. It is likely an attempt by the threat actors behind the ransomware to cut all ties to the operation and allow victims to recover and remove reasons for victims and law enforcement to pursue the threat actors any longer. It also provides plausible deniability in case the threat actors are caught with the once-secret but now public and for everyone accessible decryption keys.

References

• ¹ https://www.hornetsecurity.com/en/security-informationen-en/html-phishing-asking-for-the-password-twice/
• ² https://www.bleepingcomputer.com/news/security/synack-ransomware-releases-decryption-keys-after-el-cometa-rebrand/
• ³ https://www.bleepingcomputer.com/news/security/ragnarok-ransomware-releases-master-decryptor-after-shutdown/

Summary

In this installment of our monthly email threat review, we present an overview of the email-based threats observed in July 2021 and compare them to the previous month’s threats.

The report provides insights into:

• Unwanted emails by category
• File types used in attacks
• Industry Email Threat Index
• Attack techniques
• Impersonated company brands and organizations
• Highlighted threat email campaigns
• Ransomleaks

Unwanted emails by category

The following table shows the distribution of unwanted emails per category.

Email category	%
Rejected	84.05
Spam	11.99
Threat	3.02
AdvThreat	0.91
Content	0.03

The following time histogram shows the email volume per category per day.

The spike in rejected emails on 2021-07-30 can be attributed to the monthly reoccurring sextortion scam email campaign written in German that caused similar spikes in previous months.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain harmful content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as the sender’s identity, and the emails are not analyzed further.

File types used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicious emails)	%
Archive	30.0
PDF	20.8
HTML	17.1
Executable	11.7
Excel	7.0
Word	5.4
Other	4.3
Disk image files	3.3
Script file	0.1
Powerpoint	0.1
Email	0.0
LNK file	0.0

The following time histogram shows the email volume per file type used in attacks per 7 days.

Here we see an increase in malicious Word documents being used in attacks. In the previous month, only 3.6 % of attacks used Word documents. Word documents used to be the dominant document format used in document-based attacks. However, in 2020 many threat actors switched to Excel documents using the old Excel 4.0 macro feature to execute malicious code, causing the number of malicious Word documents used in attacks to decrease. Security vendors significantly less detected Excel 4.0 macro attacks compared to VBA macro-based attacks. However, since Microsoft has added Excel 4.0 macro support to AMSI¹ the increase in malicious Word documents could mean that threat actors may be considering switching back to using predominantly Word documents in their document-based attacks again instead of Excel documents.

Industry Email Threat Index

The following table shows our Industry Email Threat Index calculated based on the number of threat emails compared to clean emails received (in median) by each industry.

Industries	Share of threat in threat and clean emails
Research industry	4.3
Manufacturing industry	3.8
Agriculture industry	3.4
Transport industry	3.4
Education industry	3.2
Entertainment industry	3.1
Mining industry	3.1
Media industry	2.9
Hospitality industry	2.9
Healthcare industry	2.8

The following bar chart visualizes the email-based threat posed to each industry.

For comparison last month’s email-based threat index bar chart:

We observe a decrease in the overall email threat score.

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, we calculate the percent share of threat emails from each organization’s threat and clean emails to compare organizations. We then calculate the median of these percent values for all organizations within the same industry to form the industry’s final threat score.

Attack techniques

The following table shows the attack techniques used in attacks.

Attack technique	%
Other	53.2
Phishing	24.2
URL	8.1
Executable in archive/disk-image	4.7
Advance-fee scam	3.7
Extortion	2.8
Impersonation	2.2
Maldoc	1.1
LNK	0.0

The following time histogram shows the email volume per attack technique used per hour.

There are no significant anomalies.

Impersonated company brands and organizations

The following table shows which company brands and organizations our systems detected most in impersonation attacks.

Impersonated brand or organization	%
DocuSign	20.5
PayPal	13.4
Deutsche Post / DHL	12.4
Amazon	11.7
LinkedIn	4.3
Microsoft	2.5
HSBC	2.2
Santander	2.2
O2	1.7
Volks- und Raiffeisenbank	1.6

The following histogram shows the email volume for company brands and organizations detected in impersonation attacks per hour.

It’s a constant stream of phishing and other attacks impersonating big brands and organizations to entice recipients to open the emails.

Highlighted threat email campaigns

In this section, we want to highlight some malspam campaigns of prominent, well-known threat actors.

Please be advised that this does not contain all campaigns. The ranking, as well as volume figures, should therefore not be taken as a global ranking. We strive to expand this section of our reporting in the future.

The following time histogram shows the email volume for highlighted threat email campaigns per hour.

Methodology

Hornetsecurity observes thousands of different threat email campaigns of varying threat actors ranging from very unsophisticated low-effort attacks to highly complex obfuscated attack schemes. Our highlighting includes only a subset of those threat email campaigns.

Ransomleaks

Threat actors continue to leak data stolen from ransomware victims to pressure them to pay for decrypting the files and not publishing sensible data. We observed the following number of leaks on ransomware leak sites:

Leak site	Number of victim data leaks
LockBit 2.0	66
Conti	36
Hive	16
Synack	11
Lorenz	8
REvil	7
Everest	5
Promethous	5
Vice Society	4
Grief	3
RansomEXX	3
Cl0p	2
LV	2
RagnarLocker	2
Xing Team	2
Nephilim	1

The following bar chart visualizes the number of victim data leaks per leak site.

A new addition to our data set is the leak site of LockBit 2.0. The site has been up already in June, and the actors behind the site were looking for affiliates.

It offered its potentials partners in crime a performance comparison of the LockBit 2.0 ransomware compared to other RaaS (ransomware-as-a-service) offerings.

It also offered a comparison of transfer speeds for data exfiltration.

Then on 2021-07-13, the leak site started posting data of victims.

With 66 victims, LockBit 2.0 announced almost twice as many victims as Conti, the leak site with the second most victims this month.

 

References

• ¹ https://www.microsoft.com/security/blog/2021/03/03/xlm-amsi-new-runtime-defense-against-excel-4-0-macro-malware/

Summary

In this installment of our monthly email threat review, we present an overview of the email-based threats observed in June 2021 and compare them to the previous month’s threats.

The report provides insights into:

• Unwanted emails by category
• File types used in attacks
• Industry Email Threat Index
• Attack techniques
• Impersonated company brands or organizations
• Highlighted threat email campaigns
• Ransomleaks
• Special events

Unwanted emails by category

The following table shows the distribution of unwanted emails by category.

Email category	%
Rejected	84.05
Spam	11.57
Threat	3.41
AdvThreat	0.94
Content	0.03

The following time histogram shows the email volume per category per day.

Around 2021-06-13, we registered a large spike in rejected emails. Based on a significant text overlap, we can attribute this to a German-language sextortion scam campaign we observed in previous months.

As of writing, the campaign netted the criminals US$ 4,351 in BTC. Therefore the campaign is most likely profitable and thus will most likely return next month.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain harmful content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as the sender’s identity, and the emails are not analyzed further.

File types used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicious emails)	%
Archive	29.0
HTML	16.9
PDF	15.0
Other	13.0
Executable	11.0
Excel	6.9
Disk image files	4.4
Word	3.6
Powerpoint	0.1
Email	0.0
Script file	0.0
LNK file	0.0

The following time histogram shows the email volume per file type used in attacks per 7 days.

Between 2021-06-07 and 2021-06-10, Hornetsecurity detected a rise in executable email attachments. We can attribute this to a malspam campaign containing a Nanocore RAT executable dropping Agent Tesla in an archive attached to the email.

Industry Email Threat Index

The following table shows our Industry Email Threat Index calculated based on the number of threat emails compared to clean emails received (in median) by each industry.

Industries	Share of threat in threat and clean emails
Transport industry	5.6
Research industry	5.4
Entertainment industry	4.7
Education industry	4.6
Manufacturing industry	4.5
Hospitality industry	3.9
Media industry	3.8
Healthcare industry	3.7
Retail industry	3.6
Unknown	3.4

The following bar chart visualizes the email-based threat posed to each industry.

For comparison last month’s email-based threat index bar chart:

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, we calculated the percent share of threat emails from each organization’s threat and clean emails to compare organizations. We then calculate the median of these percent values overall organizations within the same industry to form the industry’s final threat score.

Attack techniques

The following table shows the attack technique used in attacks.

Attack technique	%
Other	50.8
Phishing	26.0
URL	9.4
Extortion	4.3
Executable in archive/disk-image	4.0
Advance-fee scam	2.4
Impersonation	2.2
Maldoc	1.0
LNK	0.0

The following time histogram shows the email volume per attack technique used per hour.

Between 2021-06-07 and 2021-06-10, there were elevated levels of executables in archives. This is due to the campaign, as mentioned earlier, delivering Nanocore RAT as executable in an archive file (e.g., “.7z”, “.Zip”).

Impersonated company brands or organizations

The following table shows which company brands our systems detected most in impersonation attacks.

Impersonated brand or organization	%
DocuSign	19.8
Other	15.9
Deutsche Post / DHL	15.7
Amazon	11.6
PayPal	8.6
LinkedIn	5.9
Microsoft	2.5
O2	2.1
HSBC	2.0
Santander	1.9

The following time histogram shows the email volume for company brands detected in impersonation attacks per hour.

It’s a constant stream of phishing and other attacks impersonating big brands to entice recipients to open the emails.

Starting on 2021-06-07, we observed a large-scale phishing campaign impersonating LinkedIn.

Highlighted threat email campaigns

In this section, we want to highlight some malspam campaigns of prominent, well-known threat actors.

The following time histogram shows the email volume for highlighted threat email campaigns per hour.

Please be advised that this does not contain all campaigns. The ranking, as well as volume figures, should therefore not be taken as a global ranking. We strive to expand this section of our reporting in the future.

While the Hancitor campaign overshadows all other campaigns w.r.t. to volume per hour, we can see that the QakBot malspam of botnet group tr we saw emerging last month has established itself as a reoccurring campaign. Such endless running campaigns are usually only observed by very low-quality malspam campaigns or by more sophisticated spammers such as the Emotet botnet. To this end, QakBot, as previously reported, uses email conversation thread hijacking.

Methodology

Hornetsecurity observes thousands of different threat email campaigns of varying threat actors ranging from very unsophisticated low-effort attacks to highly complex obfuscated attack schemes. Our highlighting includes only major sophisticated threat email campaigns.

Ransomleaks

Sophisticated threat actors exfiltrate confidential data from their victim’s networks. Exfiltrated data is then used as a method to pressure their victims into paying a ransom. If the victim does not pay the ransom, the confidential data is being published by the threat actors on so-called leak sites that are often only reachable through the TOR network. This trend continued in June. We observed the following number of leaks on ransomware leak sites:

Leak site	Number of victim data leaks
Conti	66
Pysa	41
REvil	28
Promethous	22
Vice Society	14
Grief	11
Avaddon	10
Lorenz	8
Everest	7
RagnarLocker	6
Xing Team	5
Cl0p	4
Synack	4
LV	3
Hive	3
Cuba	3
RansomEXX	2
Suncrypt	1
MountLocker	1

The following bar chart visualizes the number of victim data leaks per leak site.

We added data collection for the following ransomware leak sites:

The leak site of the LV ransomware:

The LV ransomware re-purposes code of the REvil ransomware. The operators don’t seem to have access to REvil’s source code but have adapted an existing REvil ransomware binary by modifying strings in its binary code.¹

The leak site of the Hive ransomware:

The Hive ransomware seems to be a new ransomware strain.

The leak site of the Vice Society ransomware:

Experts in the field of ransomware have concluded that Vice Society ransomware is identical to the HelloKitty ransomware.²

Special events

Because there have been several noteworthy events concerning the broader email threat landscape, we summarized them in this special section.

Avaddon ransomware releases decryption keys

On 2021-06-11, Avaddon released keys for over 2,934 victims.³ The Avaddon leak site listed only 186 victims that refused to pay the ransom. This means that Avaddon had 15-times more victims than published on their leak site. Under the assumption that the other ransomware operations have a similar ratio, the number of ransomware victims could be obtained by multiplying the number of victims on leak sites by 15.

Clop ransomware arrests

On 2021-06-16, the National Police of Ukraine announced they had arrested individuals suspected to have infected companies with the Clop ransomware.⁴ However, because the Clop ransomware operation continued running without interruption, it is assumed that the individuals arrested were only unimportant figures in the Clop ransomware operation, such as money mules, or sub-contractors.

We previously reported how the Clop ransomware is spread via malicious emails.

Gozi arrested

On 2021-06-29, the Office of the Attorney General of the Nation of Columbia has announced the arrest of one individual⁵ wanted by the U.S. since 2013⁶ in connection with the Gozi malware. The individual operated a bulletproof host that helped cybercriminals distribute the Gozi malware and commit other cybercrimes, such as distributing malware including the Zeus Trojan and the SpyEye Trojan, initiating and executing distributed denial of service (DDoS) attacks, and transmitting spam.

TrickBot developer arrested

On 2021-06-15, the U.S. Department of Justice announced the arrest of a 55-year-old Latvian woman on multiple charges (19 counts of a 47-count indictment) for participating in the development of the TrickBot malware.⁷

We previously reported on how TrickBot is spread via malicious emails.

 

References

• ¹ https://www.secureworks.com/research/lv-ransomware
• ² https://twitter.com/demonslay335/status/1403109032014061568
• ³ https://twitter.com/BleepinComputer/status/1403387905096175617
• ⁴ https://www.npu.gov.ua/news/kiberzlochini/kiberpolicziya-vikrila-xakerske-ugrupovannya-u-rozpovsyudzhenni-virusu-shifruvalnika-ta-nanesenni-inozemnim-kompaniyam-piv-milyarda-dolariv-zbitkiv/
• ⁵ https://www.fiscalia.gov.co/colombia/asuntos-internacionales/cae-presunto-ciberdelincuente-senalado-de-afectar-con-un-virus-informatico-a-mas-de-un-millon-de-usuarios-en-estados-unidos-y-europa/
• ⁶ https://www.justice.gov/usao-sdny/pr/three-alleged-international-cyber-criminals-responsible-creating-and-distributing-virus
• ⁷ https://www.justice.gov/opa/pr/latvian-national-charged-alleged-role-transnational-cybercrime-organization

Summary

In this installment of our monthly email threat review, we present an overview of the email-based threats observed in May 2021 and compare them to the previous month’s threats.

The report provides insights into:

• Unwanted emails by category
• Filetypes used in attacks
• Industry Email Threat Index
• Impersonated company brands or organizations
• Highlighted threat email campaigns
• Ransomleaks

Unwanted emails by category

The following table shows the distribution of unwanted emails by category.

Email category	%
Rejected	81.56
Spam	13.81
Threat	3.73
AdvThreat	0.87
Content	0.03

The following time histogram shows the email volume per category per day.

The spike in rejected emails between 2021-05-23 and 2021-05-26 is caused by a wave of sextortion scam emails targeting German language users.

This campaign and the resulting spike in rejected emails is similar to the one observed in March¹ in which the attackers made around 5,000 €. This time there were no incoming transactions to the attacker’s Bitcoin wallets used in the attack.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain harmful content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as the sender’s identity, and the emails are not analyzed further.

Filetypes used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicious emails)	%
Archive	33.1
Other	19.9
HTML	17.8
Excel	7.6
PDF	7.5
Executable	5.8
Disk image files	4.7
Word	2.8
Powerpoint	0.6
Script file	0.2
Email	0.1
LNK file	0.0

The following time histogram shows the email volume per file type used in attack per 7 days.

Industry Email Threat Index

The following table shows our Industry Email Threat Index calculated based on the number of threat emails compared to clean emails received (in median) by each industry.

Industries	Share of threat in threat and clean emails
Research industry	6.9
Transport industry	5.3
Manufacturing industry	5.0
Education industry	4.1
Healthcare industry	4.0
Media industry	3.9
Automotive industry	3.9
Entertainment industry	3.8
Utilities	3.8
Hospitality industry	3.8

The following bar chart visualizes the email-based threat posed to each industry.

For comparison last month’s email threat index bar chart:

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, we calculated the percent share of threat emails from each organization’s threat and clean emails to compare organizations. We then calculate the median of these percent values overall organizations within the same industry to form the industry’s final threat score.

Impersonated company brands or organizations

The following table shows which company brands our systems detected most in impersonation attacks.

Impersonated brand or organization	%
DocuSign	19.9
Amazon	17.6
Deutsche Post / DHL	16.8
Other	15.4
LinkedIn	4.0
Microsoft	2.8
PayPal	2.7
1&1	2.5
HSBC	2.1
O2	2.0

The following time histogram shows the email volume for company brands detected in impersonation attacks per hour.

It’s a constant stream of phishing and other attacks impersonating big brands to entice recipients to open the emails.

Highlighted threat email campaigns

In this section, we want to highlight some malspam campaigns of prominent, well-known threat actors.

The following time histogram shows the email volume for highlighted threat email campaigns per hour.

We can see that the malspam waves of the selected campaigns have well-defined start and endpoints, unlike less sophisticated mass-spam email campaigns, which will send email in a constant stream.

Please be advised that this does not contain all campaigns. The ranking, as well as volume figures, should therefore not be taken as a global ranking.

From the data, we can see that the Hancitor malspam impersonating DocuSign is a consistent reoccurring high volume malspam campaign.

At the end of the month, the increase in URL-based QakBot “tr” (named after the configuration tag/campaign ID tr found in the distributed QakBot malware’s configuration) malspam can be seen. This campaign (as other QakBot malspam campaigns) uses email conversation threat hijacking.²

Another interesting aspect of the campaign is that it uses non-clickable plain text URLs missing the http://. Victims must thus copy the URL manually to their browser. This is likely done in hopes that URL filters do not detect the URL in this form. Obviously, the campaign uses various compromised websites to host the malicious ZIP archives, further complicating detection.

Methodology

Hornetsecurity observes hundreds and thousands of different threat email campaigns of varying threat actors ranging from very unsophisticated low-effort attacks to highly complex obfuscated attack schemes. Our highlighting includes only major sophisticated threat email campaigns.

Ransomleaks

Threat actors continue to leak data stolen from ransomware victims to pressure them into paying not only for decrypting the files encrypted by the ransomware but also for not making the data stolen before encryption public. We observed the following number of leaks on ransomware leaksites:

Leaksite	Number of victim data leaks
Conti	73
Avaddon	56
Pysa	33
Darkside	25
REvil	21
Lorenz	19
Babuk	15
Xing Team	13
Nephilim	8
Cuba	6
Networm	5
Grief	5
Cl0p	4
RansomEXX	4
MountLocker	3
Everest	3
RagnarLocker	3
Astro Team	2

The following bar chart visualizes the number of victim data leaks per leaksite.

We added data collection for the following ransomware leaksites:

The leaksite of the Cuba ransomware was first seen at the end of 2019. However, the group was not very active. The leaksite currently features six leaks.

Cuba ransomware has recently been reported to be cooperating with the Hancitor malware, the malspam campaigns of which we reported on in a previous section. We, therefore, added the Cuba ransomware leaksite to our dataset.

The leaksite of the Xing Team was announced via the Astro Team leaksite on 2021-05-06 as a new partner of the Astro Team.

The Xing Team leaksite features an identical layout to the Astro Team leaksite, but contains different data leaks.

What this partnership entails is currently unknown.

Another leaksite that emerged this month is the leaksite of the Prometheus ransomware. On its site, they claim an affiliation with the REvil ransomware. However, it needs to be seen what that affiliation is and whether the REvil ransomware even knows that this new leaksite claims an affiliation with them.

The Prometheus leaksite does not remove entries for companies that have paid the ransom or for data it could sell. The respective entries are simply updated to reflect whether the company paid or the data has been soled.

Another new entry is the Grief leaksite.

Currently, it is unknown what ransomware is associated with the Grief leaksite, whether it is an already known ransomware rebranded as Grief or whether they use their own ransomware.

References

• ¹ https://www.hornetsecurity.com/en/threat-research/email-threat-review-march-2021/#unwanted-emails-by-category
• ² https://www.hornetsecurity.com/en/security-information/email-conversation-thread-hijacking/

Summary

In this second installment of our monthly email threat review, we present an overview of the email-based threats observed in April 2021 and compare it with March 2021.

The report provides insights into:

• Unwanted emails by category
• Filetypes used in attacks
• Industry Email Threat Index
• Attack techniques
• Impersonated company brands or organizations
• Highlighted threat email campaigns
• Ransomleaks

Unwanted emails by category

The following table shows the distribution of unwanted emails by categories.

Email category	%
Rejected	78.60
Spam	16.41
Threat	4.05
AdvThreat	0.89
Content	0.04

The following time histogram shows the email volume per category per hour.

Unlike in March, there were no obvious anomalous spikes in the unwanted email volumes in April.

Methodology

The listed email categories correspond to the email categories listed in the Email Live Tracking of Hornetsecurity’s Control Panel. So our users are already familiar with them. For others, the categories are:

Category	Description
Spam	These emails are unwanted and are often promotional or fraudulent. The emails are sent simultaneously to a large number of recipients.
Content	These emails have an invalid attachment. The administrators define in the Content Control module which attachments are invalid.
Threat	These emails contain dangerous content, such as malicious attachments or links, or they are sent to commit crimes, such as phishing.
AdvThreat	Advanced Threat Protection has detected a threat in these emails. The emails are used for illegal purposes and involve sophisticated technical means that can only be fended off using advanced dynamic procedures.
Rejected	Our email server rejects these emails directly during the SMTP dialog because of external characteristics, such as sender’s identity, and the emails are not analyzed further.

Filetypes used in attacks

The following table shows the distribution of file types used in attacks.

File type (used in malicous emails)	%
Archive	35.6
Other	17.6
HTML	17.3
Excel	7.6
Executable	7.0
PDF	4.8
Disk image files	4.6
Word	3.5
Powerpoint	1.3
Script file	0.5
Email	0.1
LNK file	0.0

The following time histogram shows the email volume per file type used in attacks per 7 days.

Archives (.zip, .rar, .gzip, .ace, .tar.gz, etc.) are more popular. The most prevalent use for archives in attacks is compressing the malware executable and attaching it directly to the attack email. This is done in hopes that the targeted email system is not able to scan compressed attachments. Low-quality criminal threat actors often use this technique as it does not require any technical expertise. Another use for archives is to compress malicious documents. This is also done to reduce detection.

HTML files (.htm, .html, etc.) are used either for phishing, having the phishing website attached directly to the email¹ (thus circumventing URL filters), redirecting victims to websites for malware downloads² (again to not directly include a clickable URL in the email), or social engineering.

Excel files (.xls, .xlsm, .xlsx, .xslb, etc.) with their XLM macros gained popularity last year. Unlike VBA macros malware, XLM macro malware is less detected and thus favored by many threat actors.^3,4 In fact, many threat actors use the same malicious document generator called “EtterSilent” to generate their XLM macro documents.

PDFs (.pdf) use embedded links or other social engineering lures.⁴

Attaching executables (.exe) directly to emails is the laziest approach. It is used mainly by low-quality criminal threat actors.

Disk image files (.iso, .img, etc.) are used similarly to archives.⁵ Windows can automatically mount disk image files similarly to ZIP files.

Industry Email Threat Index

The following table shows the Top 10 of our Industry Email Threat Index calculated based on the number of threat emails compared to clean emails received (in median) by each industry.

Industries	Share of threat in threat and clean emails
Research industry	5.0
Manufacturing industry	3.9
Media industry	3.6
Hospitality industry	3.6
Education industry	3.5
Healthcare industry	3.5
Automotive industry	3.3
Transport industry	3.2
Retail industry	3.0
Information technology industry	3.0

The following bar chart visualizes the email-based threat posed to each industry.

• March 2021:

• April 2021:

The global median of threat emails in threat and clean emails for organizations fell from 3.7% in March to 3.0% in April. This decrease can be seen for all industries. However, while the manufacturing industry dropped from 5.3% to 3.9% and the other industries similarity the research industry remains at 5% ratio of threat emails in their received threat and clean emails.

Methodology

Different (sized) organizations receive a different absolute number of emails. Thus, to compare organizations, we calculated the percent share of threat emails from each organization’s threat and clean emails. We then calculate the median of these percent values over all organizations within the same industry to form the industry’s final threat score.

Attack techniques

The following table shows the attack technique used in attacks.

Attack technique	%
Other	38.4
Phishing	23.4
URL	20.8
Extortion	9.0
Executable in archive/disk-image	3.4
Advance-fee scam	2.8
Impersonation	1.8
Maldoc	0.4
LNK	0.0

The following time histogram shows the email volume per attack technique used per hour.

Impersonated company brands or organizations

The following table shows which company brands our systems detected most in impersonation attacks.

Impersonated brand or organization	%
Amazon	23.0
DocuSign	21.5
Deutsche Post / DHL	11.9
PayPal	3.4
Microsoft	2.9
LinkedIn	2.8
1&1	2.6
HSBC	2.2
Unicredit	1.6
O2	1.5
Others	Rest

The following time histogram shows the email volume for company brands detected in impersonation attacks per hour.

It’s a constant stream of phishing and other attacks impersonating big brands to entice recipients to open the emails.

A reoccurring Hancitor malspam campaign dominates the impersonation of the DocuSign brand.

Highlighted threat email campaigns

In this section, we want to highlight some malspam campaigns of prominent, well-known threat actors.

The following time histogram shows the email volume per hour for a list of highlighted threat email campaigns.

Please be advised that this does not contain all campaigns. The ranking, as well as volume figures, should therefore not be taken as a global ranking. We strive to expand this section of our reporting in the future.

We can see that the malspam waves of the selected campaigns have well-defined start and endpoints, unlike less sophisticated mass-spam email campaigns, which will send email in a constant stream.

As outlined earlier in the section on impersonated company brands or organizations the reoccurring Hancitor malspam campaign pretends to be documents send via DocuSign. The campaign spikes correspond to the previously spikes in DocuSign impersonation detections.

One shown campaign stands out not because of its volume but for its low detection rate by other security solutions. The Cutwail-A botnet updated their XLSM maldocs. This time it was distributing Ursnif.

At the time of delivery, the maldocs of the campaign were detected by only 3 of 62 detections on VirusTotal.

We are aware that VirusTotal detection does not represent the real dynamic detection of the listed security products. However, because the campaign was already caught by Hornetsecurity’s Spam and Malware filter using static detection signatures, it is an apples-to-apples comparison and outlines Hornetsecurity’s detection capabilities when it comes to threat email detection.

Methodology

Hornetsecurity observes hundreds and thousands of different threat email campaigns of varying threat actors ranging from very unsophisticated low-effort attacks to highly complex obfuscated attack schemes. Our highlighting includes only major sophisticated threat email campaigns.

Ransomleaks

Threat actors continue to leak data stolen from ransomware victims to pressure them into paying not only for decrypting the files encrypted by the ransomware but also for not making the data stolen before encryption public. We observed the following number of leaks on ransomware leaksites:

Leaksite	Number of victim data leaks
Conti	41
Avaddon	40
REvil	33
Darkside	19
Babuk	17
Ragnarok	12
Astro Team	10
Cl0p	9
Everest	5
RansomEXX	2
Nephilim	1
RagnarLocker	1

The following bar chart visualizes the number of victim data leaks per leaksite.

This month the Darkside ransomware group announced that they offer insider trading information for sale. Short sellers could buy information on companies compromised by Darkside ransomware before publication on the Darkside leaksite.

After the announcement, Darkside started tagging their posted victims with their respective stock tickers.

Also, this month the Babuk ransomware announced the closing of their ransomware operation.

However, only a few days later, they clarified that they would only close the ransomware part of their operation. They will continue stealing data from victims and extorting them with the publication of the stolen data.

Conclusion

We hope you found the second installment of our monthly email threat review informative. Get back next month for more and updated email threat landscape insights.

References

• ¹ https://www.hornetsecurity.com/en/security-informationen-en/html-phishing-asking-for-the-password-twice/
• ² https://www.hornetsecurity.com/de/security-information-2/clop-clop-ta505-html-malspam-analyse/
• ³ https://www.hornetsecurity.com/en/threat-research/qakbot-distributed-by-xlsb-files/
• ⁴ https://www.hornetsecurity.com/en/threat-research/bazarloaders-elaborate-flower-shop-lure/
• ⁵ https://www.hornetsecurity.com/en/security-information/information-stealer-campaign-targeting-german-hr-contacts/