titanadmin - Page 6
by titanadmin | Apr 30, 2020 | Email Scams, Phishing & Email Spam |
A new phishing campaign has been identified that uses the Microsoft Sway file sharing service as part of a three-stage attack with the goal of obtaining the Office 365 credentials of high-level executives.
Group IB researchers identified the campaign and named it PerSwaysion, although versions of the attack have been identified that have used OneNote and SharePoint. The campaign is highly targeted and has been conducted on high-level executives at more than 150 companies. The individuals behind the campaign are believed to operate out of Nigeria and South Africa, with the earliest traces of the attacks indicating the campaign has been running since around the middle of last year.
The PerSwaysion attack starts with a spear phishing email sent to an executive in the targeted organization. The phishing emails include a PDF file attachment with no malicious code embedded. The PDF file just includes a link that the user is required to click to view the content of the file. The link directs the user to file on a Microsoft Sway page, which also requires them to click a link to view the content. Microsoft Sway allows the previewing of the document and displays the content without the user having to open the document. The document states the name of the sender – a known contact – and that individual’s email address with the message that a file has been shared for review along with a hyperlink with the text ‘Read Now’. Clicking the link directs the user to a phishing page with an Office 365 Single Sign-on login prompt.
The initial PDF file, Microsoft Sway page, and the login prompt on the phishing page are all branded with Microsoft Office 365 logos, and it is easy to see how many victims would be fooled into disclosing their credentials.
Once credentials have been obtained, they are used the same day to access the Office 365 account, email data is copied from the account, and it is then used to send further spear phishing emails to individuals in the victim’s contact list. The sent emails are then deleted from the victim’s sent folder to ensure the attack is not detected by the victim.
The emails include the sender’s name in the subject line, and since they have come from the account of a known contact, they are more likely to be opened. The lure used is simple yet effective, asking the recipient to open and review the shared document.
Many of the attacks have been conducted on individuals at companies in the financial services sector, although law firms and real estate companies have also been attacked. The majority of attacks have been conducted in the United States and Canada, United Kingdom, Netherlands, Germany, Singapore, and Hong Kong.
It is possible that the attackers continue to access the compromised emails accounts to steal sensitive data. Since the campaign targets high level executives, the email accounts are likely to contain valuable intellectual property. They could also be used for BEC scams to trick employees into making fraudulent wire transfers.
by titanadmin | Apr 29, 2020 | Email Scams, Phishing & Email Spam |
The lockdown imposed due to COVID-19 has forced employees to abandon the office and work from home, with contact maintained using communications solutions such as Skype, Slack, and Zoom. Unsurprisingly the huge increase in use of these platforms has created an opportunity for cybercriminals, who are using fake notifications from these and other communication and teleconferencing platforms as lures in phishing campaigns on remote workers.
Several campaigns have been identified that take advantage of the popularity of these platforms. One campaign has recently been identified that uses Skype branding advising users that they have pending notifications. The emails are personalized and include the Skype username and have a review button for users to click to review their notifications. These emails very closely resemble the genuine emails sent to users by Skype. The emails also appear, at first glance, to have been sent from a genuine address.
The link supplied in the email directs the recipient to a hxxps website that has Skype in the domain name. Since the connection between the browser and the website is encrypted, it will display the green padlock to show that the connection is secure, as is the case on the genuine Skype domain. The webpage includes Skype branding and the logo of the company being targeted and states that the webpage has been set up for authorized use by employees of the company. The username of the victim is automatically added to the login page, so all that is required is for a password to be entered.
This campaign was identified by Cofense, which received multiple reports from business users about the emails, which bypassed Microsoft Exchange Online Protection (EOP) and were delivered to Office 365 inboxes.
A Zoom campaign has also been identified that uses similar tactics. Zoom is one of the most popular lockdown teleconferencing apps and has been recommended by many businesses for use by employees to maintain contact during the lockdown. The platform has also proven popular with consumers and now has more than 300 million users.
In this campaign, Zoom meeting notifications are sent to targets. As is common with phishing campaigns, the attackers generate fear and urgency to get the targets to respond quickly without scrutinizing the messages. This campaign advises the recipients to login to a meeting with their HR department regarding their job termination. Clicking the link will similarly direct users to a fake login page where they are required to enter their credentials. The landing page is a virtual carbon copy of the official Zoom login page, although the only parts of the page that work are the username and password fields. This campaign was identified by Abnormal Security, which reports that around 50,000 of these messages were delivered to Office 365 accounts and bypassed EOP.
The phishing emails are credible, the webpages that users are directed to look genuine, and many people will be fooled by the emails. Security awareness training will help to condition employees to question emails such as these, but given the number of messages that are bypassing Microsoft’s EOP, businesses should also consider adding an additional layer of email security to their Office 365 accounts.
This is an area where TitanHQ can help. SpamTitan Cloud does not replace EOP for Office 365, it allows businesses to add an extra layer of protection on top to provide extra protection from zero-day attacks. SpamTitan Cloud blocks spam, phishing, and malware laced emails that would otherwise be delivered to Office 365 inboxes.
SpamTitan Cloud is quick and easy to implement and can protect your Office 365 accounts in a matter of minutes. Since the solution is available on a free trial, you will be able to evaluate the difference it makes and see how many malicious messages it blocks before committing to a purchase.
For further information on improving your phishing defenses, give the TitanHQ team a call today.
by titanadmin | Apr 27, 2020 | Email Scams, Phishing & Email Spam |
Higher education institutions in the United States are being targeted in a phishing campaign that distributes a remote access trojan called Hupigon, a RAT that was first identified in 2010.
The Hupigon RAT has previously been used by advanced persistent threat groups (APT) from China, although this campaign is not believed to have been conducted by APT groups, instead the Hupigon RAT has been repurposed by cybercriminals. While several industries have been targeted in the campaign, almost half of attacks have been on colleges and universities.
The Hupigon RAT allows the operators to download other malware variants, steal passwords, and gain access to the microphone and webcam. Infection could see the attackers take full control of an infected device.
The campaign uses online dating lures to get users to install the Trojan. The emails show two dating profiles of supposed users of the platform, and the recipient is asked to select the one they find the most attractive. When the user makes their choice, they are directed to a website where an executable file is downloaded, which installs the Hupigon RAT.
The choice of lure for the campaign is no doubt influenced by the huge rise in popularity of dating apps during the COVID-19 pandemic. While there are not many actual dates taking place due to lockdown and social distancing measures now in place around the globe, the lockdowns have seen many people with a lot of time on their hands. That, coupled with social isolation for many singles, has actually led to an uptick in the use of online dating apps, with many users of the apps turning to Zoom and FaceTime to have virtual dates. Several popular dating apps have reported an increase in use during the COVID-19 pandemic. For example, Tinder reports use has increased, with the platform having its busiest ever day, with more than 3 billion profiles swiped in a single day.
As we have already seen with COVID-19 lures in phishing attacks, which account for the majority of lures during the pandemic, when there is interest in a particular event or news story, cybercriminals will take advantage. With the popularity of dating apps soaring, we can expect to see an increase in the number of online dating -themed lures.
The advice for higher education institutions and businesses is to ensure that an advanced spam filtering solution is in place to block the malicious messages and ensure they do not reach end users’ inboxes. It is also important to ensure that security awareness training continues to be provided to staff, students, and remote employees to teach them how to recognize the signs of phishing and other email threats.
TitanHQ can help with the former. If you want to better protect staff, students, and employees and keep inboxes free of threats, give the TitanHQ team a call today. After signing up, you can be protecting your inboxes in a matter of minutes.
by titanadmin | Apr 22, 2020 | Email Scams, Phishing & Email Spam |
Healthcare providers are being targeted by cybercriminals using COVID-19-themed phishing emails, with the campaigns showing no sign of letting up. The volume of attacks has prompted the U.S. Federal Bureau of Investigation (FBI) to issue a further warning to healthcare providers urging them to take steps to protect their networks and block the attacks.
The first major COVID-19-themed phishing attacks targeting healthcare providers started to be detected by around March 18, 2020. The attacks have grown over the following weeks and the lures have diversified.
Campaigns have been conducted targeting at-home healthcare employees who are providing telehealth services to patients, and there has been an increase in business email compromise scams. The latter see vendors impersonated and requests sent for early or out-of-band payments due to difficulties that are being experienced due to COVID-19.
The phishing attacks are being conducted to obtain login credentials and to spread malware, both of which are used to gain a foothold in healthcare networks to allow follow-on system exploitation, persistence, and the exfiltration of sensitive data.
The malware being distributed in these campaigns is highly varied and includes information stealers such as Lokibot, backdoors, and Trojans such as Trickbot. Microsoft has recently reported that Trickbot accounts for the majority of COVID-19 phishing emails targeting Office 365 users, with a campaign last week involving hundreds of different, unique macro-laced documents. In addition to being a dangerous malware variant in its own right, Trickbot also downloads other malicious payloads, including RYUK ransomware.
A diverse range of malware is delivered by a similarly diverse range of email attachments and malicious scripts. Microsoft Word documents containing malicious macros are commonly used, as are 7-zip compressed files, Microsoft executables, and JavaScript and Visual Basic scripts. The emails are being sent from a combination of domestic and international IP addresses.
While the number of COVID-19-themed phishing emails has been increasing, the overall volume of phishing emails has not increased by a major amount. What is happening is threat actors are changing their lures and are now using COVID-19 lures as they are more likely to be opened.
The campaigns can be highly convincing. The lures and requests are plausible, many of the emails are well written, and authorities on COVID-19 such as the Centers for Disease Control and Prevention, the HHS’ Centers for Medicare and Medicaid Services, and the World Health Organization have been spoofed. Oftentimes the emails are sent from a known individual and trusted contact, which makes it more likely that the email attachment will be opened.
The advice offered by the FBI is to follow cybersecurity best practices such as never opening unsolicited email attachments, regardless of who appears to have sent the email. Ensuring software is kept up to date and patches are applied promptly is also important, as is turning off automatic email attachment downloads. The FBI has also recommended filtering out certain types of attachments through email security software, something that is easy to do with SpamTitan.
The FBI has stressed the importance of not opening email attachments, even if antivirus software says that the file is clean. As the Trickbot campaign shows, new variants of malicious documents and scripts are being created at an incredible rate, and signature-based detection methods cannot keep up. This is another area where SpamTitan can help. In addition to using dual antivirus engines to identify known malware variants faster, SpamTitan includes sandboxing to identify and block zero-day malware threats that have yet to have their signatures added to antivirus software virus definitions lists.
Training is important to teach healthcare employees cybersecurity best practices to help them identify phishing emails, but it is also important to ensure that your technical controls are capable of blocking these threats.
by titanadmin | Apr 15, 2020 | Email Scams, Phishing & Email Spam, Spam News |
Data obtained by the UK think tank Parliament Street has revealed the extent to which universities are being targeted by cybercriminals and the sheer number of spam and malicious emails that are sent to the inboxes of university staff and students.
Data on malicious and spam email volume was obtained by Parliament Street through a Freedom of Information request. The analysis of data from UK universities showed they are having to block millions of spam emails, hundreds of thousands of phishing emails, and tens of thousands of malware-laced emails every year.
Warwick University’s figures show that more than 7.6 million spam emails were sent to the email accounts of staff and students in the final quarter of 2019 alone, which included 404,000 phishing emails and more than 10,000 emails containing malware.
It was a similar story at Bristol University, which received more than 7 million spam emails over the same period, 76,300 of which contained malware. Data from the London School of Hygiene and Tropical Medicine revealed more than 6.3 million spam emails were received in 2019, which included almost 99,000 phishing emails and more than 73,500 malware attacks. 12,773,735 spam and malicious emails were received in total for 2018 and 2019.
Data from Lancaster University revealed more than 57 million emails were rejected for reasons such as spam, malware, or phishing, with 1 million emails marked as suspected spam. The figures from Imperial College London were also high, with almost 40 million emails blocked in 2019.
Like attacks on companies, cyberattacks on universities are often conducted for financial gain. These attacks attempt to deliver malware and obtain credentials to gain access to university networks to steal data to sell on the black market. Universities store huge amounts of sensitive student data, which is extremely valuable to hackers as it can be used for identity theft and other types of fraud. Attacks are also conducted to deliver ransomware to extort money from universities.
Universities typically have high bandwidth to support tens of thousands of students and staff. Attacks are conducted to hijack devices and add them to botnets to conduct a range of cyberattacks on other targets. Email accounts are being hijacked and used to conduct spear phishing attacks on other targets.
Nation state-sponsored advanced persistent threat (APT) groups are targeting universities to gain access to intellectual property and research data. Universities conduct cutting-edge research and that information is extremely valuable to companies that can use the research data to develop products to gain a significant competitive advantage.
Universities are seen as relatively soft targets compared to organizations of a similar size. Cybersecurity defenses tend to be far less advanced, and the sprawling networks and number of devices used by staff and students make defending networks difficult.
With the number of cyberattacks on universities growing, leaders of higher education institutions need to take steps to improve cybersecurity and prevent the attacks from succeeding.
The majority of threats are delivered via email, so advanced email security defenses are essential, and that is an area where TitanHQ can help.
Independent tests show SpamTitan blocks in excess of 99.97% of spam email, helping to keep inboxes free of junk email. SpamTitan incorporates dual anti-virus engines to block known threats, machine learning to identify new types of phishing attacks, and sandboxing to detect and block zero-day malware and ransomware threats. When email attachments pass initial tests, suspicious attachments are sent to the sandbox for in-depth analysis to identify command and control center callbacks and other malicious actions. SpamTitan also incorporates SPF and DMARC controls to block email impersonation attacks, data loss prevention controls for outbound messages, and controls to detect potential email account compromises.
If you want to improve your cybersecurity defenses, start by upgrading your email security defenses with SpamTitan. You may be surprised to discover the little investment is required to significantly improve your email security defenses. For more information, call the TitanHQ team today.
by titanadmin | Apr 10, 2020 | Internet Security, Network Security |
Security awareness for remote workers has never been more important. It is fair to say that there have never been more people working from home as there are now during the COVID-19 pandemic, and home workers are now being actively targeted by cybercriminals who see them as providing an easy way to gain access to their corporate networks to steal sensitive information, and install malware and ransomware.
Businesses may have already given their employees security awareness training to make sure they are made aware of the risks that they are likely to encounter and to teach them how to recognize threats and respond. However, working from home introduces many more risks and those risks may not have been covered in security awareness training sessions geared toward protecting office workers. It is also important to provide security training for employees, and this is especially important for remote workers, as risk increases when employees are working remotely.
In this post we will highlight some of the key areas that must be addressed in work-from-home (WFH) security awareness training for the workforce.
Increased Security Awareness for Remote Workers Required as COVID-19 Crisis Deepens
Naturally, as an email security solution provider, we strongly advocate the use of a powerful email security solution and layered technical defenses to protect against phishing, but technical controls, while effective, will not stop all threats from reaching inboxes. It is all too easy to place too much reliance on technical security solutions for securing email environments and work computers. The truth is that even with the best possible email security defenses in place, some threats will end up reaching inboxes.
The importance of providing security awareness training to the workforce and the benefits of doing so have been highlighted by several studies. One benchmarking study, conducted by the security awareness training provider KnowBe4, revealed 37.9% of employees fail phishing tests if they are not provided with security awareness and social engineering training. That figure has increased by 8.3% from the previous year. With security awareness training and phishing email simulations, the figure dropped to 14.1% after 90 days.
During the COVID-19 pandemic, the volume of phishing emails being sent has increased significantly and campaigns are being conducted targeting remote workers. The aim of the phishing campaigns is to obtain login credentials to email accounts, VPNs, and SaaS platforms and to spread malware and ransomware.
With so many employees now working from home, and the speed at which companies have had to transition from a largely office-based workforce to having virtually everyone working from home may have seen security awareness training for remote workers put on the back burner. However, with the lockdown likely to be extended for several months and attacks on the rise, it is important to make sure that training is provided, and as soon as possible.
Increase in COVID-19 Domain Registrations and Rise in Web-Based Attacks
Security awareness training for remote workers also needs to cover internet security as not all threats will arrive in inboxes. Most phishing attacks have a web-based component, and malicious websites are being set up for drive-by malware downloads. Currently, the vast majority of threats are using COVID-19 and the Novel Coronavirus as a lure to get remote workers to download malware, ransomware, or part with their login credentials.
Unsurprisingly, cybercriminals have increased web-based attacks, which are being conducted using a plethora of COVID-19 and novel coronavirus-themed domains. By the end of March, approximately 42,000 domains related to COVID-19 and coronavirus had been registered. An analysis by Check Point Research revealed those domains were 50% more likely to be malicious than other domains registered over the same period.
It is important to raise awareness of the risks of using corporate laptops for personal use such as browsing the Internet. Steps should also be taken to limit the websites that can be accessed by employees and, at the very least, a solution should be implemented and configured to block access to known malicious websites that are used for phishing, fraud, and malware distribution.
Shadow IT is a Major Security Risk
When employees are office-based and connected to the network, identifying shadow IT – unauthorized software and hardware used by employees – is more straightforward. The problem not only becomes harder to identify when employees work from home, the risk of unauthorized software being loaded onto corporate-issued devices increases.
Software downloaded onto work computers carries a risk of a malware infection and potentially offers an easy way to attack the user’s device and the corporate network. IT teams will have little visibility into the unauthorized software on users’ devices and whether it is running the latest version and has been patched against known vulnerabilities. It is important to cover shadow IT in security awareness training for remote workers and to make it clear that no software should be installed on work devices and that personal USB devices should not be connected to corporate devices without the go-ahead being given by the IT department.
The COVID-19 pandemic has seen many workers turn to teleconferencing platforms to communicate with the office, friends, and family. One of the most popular teleconferencing platforms is Zoom. Malicious installers have been identified that install the genuine Zoom client but have been bundled with malware. Installers have been identified that also install adware, Remote Access Trojans, and cryptocurrency miners.
How TitanHQ Can Help Improve Email Security
Several security awareness training firms have made resources available to businesses free of charge during the COVID-19 crisis to help them train the workforce, such as the SANS Institute. Take advantage of these resources and push them out to your workforce. If you are a small SMB, you may also be able to get access to free phishing simulation emails to test the workforce and reinforce training.
TitanHQ can’t help you with your remote worker cybersecurity awareness training, but we can help by ensuring employees have to deal with fewer threats by protecting against email and web-based attacks.
SpamTitan is an advanced and powerful cloud-based email security solution that will protect remote workers from phishing, spear phishing, malware, virus, and ransomware attacks by blocking attacks at the source and preventing the threats from reaching inboxes. SpamTitan features dual anti-virus engines to protect against known malware threats and email sandboxing to block unknown (zero-day) malware threats. SpamTitan incorporates several real-time threat intelligence feeds to block current and emerging phishing attacks and machine learning technology detects and blocks previously unseen phishing threats. SpamTitan has been developed to work seamlessly with Office 365 to allow businesses to create layered defenses, augmenting Microsoft’s protections and adding advanced threat detection and blocking capabilities.
WebTitan is a DNS filtering solution that will protect all workers from web-based attacks, no matter where they access the internet. WebTitan incorporates zero-minute threat intelligence and blocks malicious domains and web pages as soon as they are identified. The solution can also be used to carefully control the types of websites that remote workers can access on their corporate-owned devices, via keyword and category-based controls. WebTitan can also be configured to block the downloading of malicious files and software installers to control shadow IT.
For more information on protecting your business during the COVID-19 crisis, to arrange a product demonstration of SpamTitan and/or WebTitan, and to register for a free trial of either solution to allow you to start instantly protecting against email and web-based threats, contact TitanHQ today!
by titanadmin | Apr 1, 2020 | Industry News |
Blackpoint Cyber announced its Remote Reality LIVE conference, which will occur online April 8th and April 9th 2020.
The conference will focus on managed service providers (MSPs) and how they can stay secure, profitable, and resilient as the world increases remote operations during the COVID-19 pandemic – registration and attendance are free. The two-day conference will include sessions by former leaders of the United States’ government cyber security and intelligence communities as well as cyber security experts and business veterans from the MSP services and technology industry.
Blackpoint Cyber announces its virtual cyber security conference for MSPs – Remote Reality LIVE. Featuring a keynote from the former Acting Director of the CIA and sessions from tech giants Datto, Webroot, Marketopia, and more.
Jon Murchison, Blackpoint’s CEO and founder, and former US government cyber operations expert, explains the conference’s objective: “IT services and infrastructure have become mission critical for organizations to survive in this new economic landscape brought on by COVID-19. MSPs are the key to our success and, especially during these times, a collective national asset to their respective countries. That’s why we are bringing together experienced government and industry leaders to help MSPs navigate the current economic and security environments. We’re excited to provide one of the first online and socially-distanced conferences dedicated to MSPs and cyber security.”

Blackpoint has partnered with leading technology, service, and marketing firms for the conference, including:
- Datto: leading global provider of cloud-based software and technology solutions purpose-built for MSPs
- Webroot: Cybersecurity Solutions Purpose-Built for MSPs and SMBs
- Convergint: Global, Service-based Systems Integrator
- Marketopia: Lead Generation and Marketing for Technology Companies
- ID Agent: Dark Web and Identity Theft Protection
- TitanHQ: Email and DNS Security
- Compliancy Group: HIPAA Compliance-as-a-Service
- Atlantic Data Forensics: Premier Incident Response and Forensics
- ProSource Technology Solutions: Leading Managed Service Provider
- Corporate Office Properties Trust (COPT): Premier Real Estate Investment Trust
Michael Morell, former Deputy Director and Acting Director CIA, will present the keynote session on national security implications of the Coronavirus outbreak. While at the CIA, Mr. Morell was President George W. Bush’s daily intelligence briefer during the 9/11 attacks and was awarded the Distinguished Intelligence Medal, the CIA’s second highest honor.
Additional former US government cyber security and intelligence expert speakers include: Bill Priestap, former FBI Assistant Director of Counterintelligence, Chris Inglis, Former Deputy Director of NSA, Dave Sears, retired Commander and Navy SEAL, and Kevin Donegan, former United States Navy Vice Admiral and previous commander of the US Navy’s 5th fleet out of Bahrain. Security and MSP industry leaders will also present informational sessions, such as lead generation in a virtual world, security in the MSP space, cyber security for commercial real estate, the threat landscape of remote workers, and more.
Matt Solomon, VP of Business Development & IT at ID Agent, shares his sentiments on the conference: “ID Agent is very excited to participate in one of the first virtual MSP events since in-person events have been taken off the schedule. MSPs still need education during this period and we are honored to be part of such an esteemed group of vendors.”
In addition to learning how to stay secure and prosper, conference attendees will also be eligible for giveaways and prizes.
Participants may register online: Remote Reality Live – Free Registration

by titanadmin | Mar 26, 2020 | Internet Security, Network Security, Spam Software, Website Filtering |
When it comes to cybersecurity and home working, CIOs and IT teams have a challenge – How to ensure the same level of protection is provided for remote workers as they get when they are in the office. To help we have compiled a set of cybersecurity best practices for home workers to help IT teams prepare for a massive increase in telecommuting
The cybersecurity protections at home will not be nearly as good for home workers as protections in the office, which are much easier to implement and maintain. IT departments will therefore need to teach telecommuting workers cybersecurity best practices for home working and their devices will need to be configured to access applications and work resources securely. With so many workers having to telecommute, this will be a major challenge.
The coronavirus pandemic has forced businesses to rapidly expand the number of telecommuting workers and having to increase capacity in such a short space of time increases the potential for mistakes. Further, testing may not be nearly as stringent as necessary given the time pressure IT workers are under. Their teams too are likely to be depleted due to self-isolating workers.
One area where standards are likely to slip is staff training on IT. Many employees will be working from home for the first time and will have to use new methods and applications they will not be familiar with. The lack of familiarity can easily lead to mistakes being made. It is important that even though resources are limited you still teach cybersecurity best practices for home workers. Do not assume that telecommuting workers will be aware of the steps they must take to work securely away from the office.
Steps for IT Teams to Take to Improve Cybersecurity for Home Workers
Listed below are some of the key steps that IT teams need to take to improve security for employees that must now work from home.
Ensure VPNs are Provided and Updated
Telecommuting workers should not be able to access their work environment unless they use a VPN. A VPN will ensure that all traffic is encrypted, and data cannot be intercepted in transit. Enterprise-grade VPNs should be used as they are more robust and provide greater security. Ensure there are sufficient licenses for all workers, and you have sufficient bandwidth available. You must also make sure that the VPN is running the latest software version and patches are applied, even if this means some downtime to perform the updates. VPN vulnerabilities are under active attack.
Set up Firewalls for Remote Workers
You will have a firewall in place at the office and remote workers must have similar protections in place. Software firewalls should be implemented to protect remote workers’ devices. Home routers may have inbuilt firewalls. Talk employees through activating hardware firewalls if they have them on their home routers and ensure that passwords are set to prevent unauthorized individuals from connecting to their home Wi-Fi network.
Apply the Rule of Least Privilege
Remote workers introduce new risks, and with large sections of the workforce telecommuting, that risk is considerable. Remote workers are being targeted by cybercriminals and through web- and email-based attacks. In the event of a malware infection or credential theft, damage can be limited by ensuring workers only have access to resources absolutely necessary for them to perform their work duties. If possible, restrict access to sensitive systems and data.
Ensure Strong Passwords are Being Set
To protect against brute force attacks, ensure good password practices are being followed. Consider using a password manager to help employees remember their passwords. The use of complex passwords should be enforced.
Implement Multifactor Authentication
Multifactor authentication should be implemented on all applications that are accessed by remote workers. This measure will ensure that if credentials are compromised, system access is not granted unless a second factor is provided.
Ensure Remote Workers’ Devices Have Antivirus Software installed
Antivirus software must be installed on all devices that are allowed to connect to work networks and the solutions must be set to update automatically.
Set Windows Updates to Automatic
Working remotely makes it harder to monitor user devices and perform updates. Ensure that Windows updates are set to occur automatically outside of office hours. Instruct workers to leave their devices on to allow updates to take place.
Use Cloud-Based Backup Solutions
To prevent accidental data loss and to protect against ransomware attacks, all data must be backed up. By using cloud-based backups, in the event of data loss, data can be restored from the cloud-backup service.
Teach Cybersecurity Best Practices for Home Workers
All telecommuting workers must be shown how they need to access their work environment securely when working away from the office. Reinforce IT best practices with home workers, provide training on the use of VPNs, provide training on cybersecurity dos and don’ts when working remotely, and explain procedures for reporting problems.
Define Procedures for Dealing with a Security Incident
Members of the IT team are also likely to be working remotely so it is essential that everyone is aware of their role and responsibilities. In the event of a security incident, workers should have clear procedures to follow to ensure the incident is resolved quickly and efficiently.
Implement a Web Filter
A web filter will help to protect against web-based malware attacks by blocking access to malicious websites and will help to prevent malware downloads and the installation of shadow IT. Also consider applying content controls to limit employee activities on corporate-owned devices. Drive-by malware attacks have increased and the number of malicious domains registered in the past few weeks has skyrocketed.
Use Encrypted Communication Channels
When you need to communicate with telecommuting workers, ensure you have secure communications channels to use where sensitive information cannot be intercepted. Use encryption for email and secure text message communications, such as Telegram or WhatsApp.
Ensure Your Email Security Controls are Sufficient
One of the most important cybersecurity best practices for home workers is to take extra care when opening emails. Phishing and email-based malware attacks have increased significantly during the coronavirus pandemic. Ensure training is provided to help employees identify phishing emails and other email threats.
Consider augmenting email security to ensure more threats are blocked. If you use Office 365, a third-party email security solution layered on top will provide much better protection. Exchange Online Protection (EOP) is unlikely to provide the level of protection you need against phishing and zero-day malware threats. Consider an email security solutions with data loss protection functions to protect against insider threats.
Monitor for Unauthorized Access
More devices connecting to work environments makes it much easier for threat actors to hide malicious activity. Make sure monitoring is stepped up. An intrusion detection system that can identify anomalous user behavior would be a wide investment.
For further information on enhancing email security and web filtering to protect remote workers during the coronavirus pandemic, contact TitanHQ today.
by titanadmin | Mar 24, 2020 | Phishing & Email Spam, Spam Software |
In this post, we explore email security and home working and offer advice to help businesses ensure their workers, devices, and networks are protected.
The 2019 Novel Coronavirus pandemic has forced many workers to self-isolate at home and an increasing number of employees want to work from home to reduce the risk of contracting COVID-19. Businesses are under pressure to allow their workers to stay at home and use either company-issued or personal devices to access their networks and work remotely.
Cybercriminals are constantly changing their tactics, techniques, and procedures and they have jumped at the opportunity provided by the Novel Coronavirus. People are scared and rightly so. COVID-19 has a high mortality rate and the virus is spreading like wildfire. People want information about cases in their local area, advice on how to protect themselves, and information about possible cures. Cybercriminals have obliged and are conducting phishing campaigns that claim to offer all that information. Many campaigns have now been detected from many different threat groups that attempt to obtain login credentials and spread malware. Since early January when the first major campaigns were detected, the volume of coronavirus and COVID-19 emails has increased significantly.
Campaigns are being conducted impersonating authorities on the Novel Coronavirus and COVID-19, such as the World Health Organization (WHO), the U.S. Centers for Disease Control and Prevention (CDC), the U.S. Department of Health and Human Services, and other government agencies. COVID-19-themed emails are being sent to remote workers that spoof HR departments warning about cases that have been detected within the organization. Health insurers are being spoofed in campaigns that include invoices for coverage for COVID-19.
Since January, more than 16,000 Coronavirus and COVID-19-themed domains have been registered which are being used to host phishing kits and distribute malware. Researchers at CheckPoint Software report that those domains are 50% more likely to be malicious than other domains registered in the same period.
Email security and home working will naturally be a major concern for IT teams given the sheer number of home workers due to the Coronavirus pandemic and the volume of attacks that are now being conducted targeting home workers. With so many devices now connecting to networks remotely, if cybercriminals do obtain credentials, it will be much harder for IT teams to identify threat actors connecting remotely. Fortunately, there are steps that can be taken to improve email security and home working need not majorly increase risk.
You should make sure that your employees can only connect to your network and cloud-based services through a VPN. Enterprise VPNs can be configured to force all traffic through the VPN to reduce the potential for error. Make sure that the VPN is configured to start automatically when the device is powered up.
It is crucial that all remote workers are protected by a robust and effective email security solution. It is not possible to stop cybercriminals targeting remote workers, but it is possible to stop phishing and malware threats from reaching inboxes.
To protect your employees against phishing attacks and malware, an advanced email security solution is essential. If you use Office 365 for email, do not rely on Office 365 email security. You will need greater protection than Exchange Online Protection provides to protect against phishing, spear phishing, and zero-day threats.
SpamTitan has multiple detection mechanisms to identify and block the full range of email threats. SpamTitan incorporates SPF and DMARC to provide protection against email impersonation attacks, machine learning algorithms and predictive technology to protect against zero-day attacks, advanced phishing protection from whaling and spear phishing attacks by scanning inbound email in real-time, dual antivirus engines to block malware threats, and email sandboxing for in-depth analysis of suspicious attachments. SpamTitan also includes 6 specialist RBLs, supports whitelisting, blacklisting, and greylisting, and incorporates multiple threat intelligence feeds.
There is an increased risk of insider threats with remote workers. To provide protection and prevent accidental policy violations, SpamTitan incorporates a data loss prevention filter to stop credit card numbers, Social Security numbers, and other data types from being sent via email.
No email security solution will be able to block 100% of email threats, 100% of the time. It is therefore important to provide regular cybersecurity training to employees to make them aware of phishing threats, train them how to identify a phishing email or social engineering scam, and to condition remote employees how to respond should a threat be received. Phishing simulation exercises are also useful to find out which employees require additional training and to identify possible gaps in training programs. IT security basic training refreshers should also be provided to ensure employees know what can and cannot be done with work devices.
Multifactor authentication must be implemented on all applications and email accounts to provide protection in the event of an account compromise. If credentials are stolen and used from a previously unknown location or an unfamiliar device, a second authentication factor must be provided before access is granted. You should also disable macros on all user devices unless a specific user needs to use macros for work.
You can arrange a demonstration to see SpamTitan in action and you can also sign up for a free trial to put SpamTitan to the test in your own environment.
by titanadmin | Mar 19, 2020 | Phishing & Email Spam, Spam Software |
The TrickBot Trojan is a sophisticated banking Trojan that was first identified in 2016. While the malware was initially just an information stealer concerned with stealing online banking credentials, the malware has evolved considerably over the past four years and several modules have been added that provide a host of additional malicious capabilities.
The TrickBot Trojan’s information stealing capabilities have been significantly enhanced. In addition to banking credentials, it will steal system and network information, email credentials, tax data, and intellectual property. TrickBot is capable of moving laterally and silently infecting other computers on the network using legitimate Windows utilities and the EternalRomance exploit for the SMBv1 vulnerability. The malware can add a backdoor for persistent access. TrickBot also serves as a malware downloader and will download other malicious payloads, including Ryuk ransomware.
The Trojan is frequently updated and new variants are regularly released. The Command and Control infrastructure is also constantly changing. According to an analysis by Bitdefender, more than 100 new IPs are added to its C&C infrastructure each month with each having a lifespan of around 16 days. The malware and its infrastructure are highly sophisticated, and while steps have been taken to dismantle the operation, the attackers are managing to stay one step ahead.
TrickBot is primarily distributed by spam email through the Emotet botnet. Infection with Emotet sees TrickBot downloaded, and infection with TrickBot sees a computer added to the Emotet botnet. Once all useful information has been obtained from an infected system, the baton is passed over to the Ryuk ransomware operators with a reverse shell opened giving the Ryuk ransomware operators access to the system.
A recent analysis of a variant captured by Bitdefender on January 30, 2020 has shown another method of distribution has been added to its arsenal. The Trojan now has a module for bruteforcing RDP. The brute force RDP attacks are mainly being conducted on organizations in the financial services, education, and telecom industries and are currently targeted on organizations in the United States and Hong Kong at this stage, although it is likely that the attacks will spread geographically over the coming weeks. The attacks are being conducted to steal intellectual property and financial information.
Since the TrickBot Trojan is modular, it can be constantly updated with new features and the evolution of the malware so far, and its success, means it will continue to be a threat for some time to come. Fortunately, it is possible to prevent infections by practicing good cyber hygiene.
Spam is still the primary method of delivery for both the Emotet Trojan and TrickBot so an advanced spam filter is essential. Since new variants are constantly being released, signature-based detection methods alone are insufficient. SpamTitan incorporates a Bitdefender-powered sandbox to analyze suspicious email attachments for malicious activity. This ensures the malicious activity of never-before-seen malware variants is identified and the emails are quarantined before they can cause any harm.
If you don’t need RDP, ensure it is disabled. If you do, ensure access is restricted and strong passwords are set. Use rate limiting to block login attempts after a set number of failures and ensure multifactor authentication is implemented to stop stolen credentials from being used.
For further information on SpamTitan Email Security and to find out how you can improve your defenses against email and web-based attacks, contact the TitanHQ team today.
by titanadmin | Mar 18, 2020 | Email Scams, Phishing & Email Spam |
The City of Durham and the County of Durham in North Carolina have experienced a ransomware attack that has crippled both. The attack ‘started’ on March 6 in the late evening, which is common for ransomware attacks. Most take place in the evening and over the weekend, when there is less chance of the file encryption being detected.
Two separate attacks occurred simultaneously. Fast action by the IT department helped to contain the attack, but not in time to prevent approximately 80 servers from being infected. Those servers were encrypted and need to be rebuilt and approximately 1,000 computers had to be re-imaged.
There are many ways that cybercriminals gain access to business networks to deploy malware, but email is the most common attack vector. Most cyberattacks start with a phishing email and this attack was no different.
Ryuk ransomware was used to encrypt files on the network in order to extort money from the city and country. A ransom demand is issued which, depending on the extent of encryption, can range from several thousand dollars to several million. This phase of the attack is the most visible and causes the most disruption, but the attack actually started much earlier.
Ruyk ransomware is delivered by the TrickBot Trojan, an information stealer turned malware downloader. One installed on a networked device, the TrickBot Trojan performs reconnaissance, moves laterally, and installs itself on other computers on the network. Once all useful information has been found and exfiltrated, a reverse shell is opened and access to the system is given the ransomware operators. They will then move laterally and download their ransomware payload onto as many devices as possible on the network.
TrickBot downloaded by Emotet malware, a notorious botnet and Emotet is delivered via email. The Emotet campaigns used a combination of Office documents with malicious macros that download the malware payload and hyperlinks to websites where malware is downloaded. TrickBot may also be delivered directly through spam email. This Trio of malware variants can do a considerable amount of damage. Even if the ransom is not paid, losses can be considerable. The Trojans can steal a substantial amount of sensitive information including email credentials, banking credentials, tax information, and intellectual property.
In this case, seven computers appear to have been compromised in the first phase of the attack as a result of employees responding to phishing emails.
The key to blocking attacks such as this is to have layered defenses in place that are capable of blocking the initial attack. That means an advanced spam filtering solution is required to block the initial phishing emails and end users must receive regular security awareness training to help them identify any malicious emails that arrive in their inboxes. Multifactor authentication is needed to prevent stolen credentials from being used to access email accounts and endpoint security solutions are required to detect malware if it is downloaded.
To find out more about protecting your systems from phishing and malware attacks, and how a small per user cost per month can prevent a hugely expensive cyberattack, give the TitanHQ team a call today.
by titanadmin | Mar 12, 2020 | Email Scams, Phishing & Email Spam, Spam Advice, Spam News, Spam Software, Website Filtering |
Several new COVID-19 phishing email campaigns have been detected over the past few days that are exploiting fear about the novel coronavirus pandemic to deliver computer viruses and steal sensitive information.
People are naturally worried about getting infected with the real virus especially with the high fatality rate, so emails related to COVID-19 are likely to be opened.
Some of the phishing emails that have been intercepted are easy to identify as malicious. They are poorly written with spelling mistakes and grammatical errors, but some campaigns have been expertly crafted and are highly convincing and are likely to catch out many people.
The first COVID-19 phishing campaigns were detected in January and the number has steadily grown over the past few weeks. Many different threat groups are now using COVID-19 phishing lures to fool the unwary into disclosing credentials, visiting malicious links, or downloading malware.
The World Health Organization (WHO) has issued a warning after several phishing campaigns were detected that impersonated WHO. The emails claimed to provide essential information about cases in the local area along with advice on how to avoid infection. One of the most recently detected campaigns claimed to provide “Coronavirus Updates” with the emails containing a ZIP file attachment that appeared to be a PDF file – MYHEALTH.PDF. However, the file was actually an executable file – MYHEALTH.exe. If the file was opened, it triggered the download of GULoader, which in turn downloads Formbook malware from Google Drive. Another similar campaign included a Word attachment that downloaded the TrickBot Trojan, which is being used to deliver Ryuk ransomware as a secondary payload.
The Centers for Disease Control and prevention is also being impersonated. One campaign claims the novel coronavirus had become an airborne threat and warns of new cases in the local area. The emails appear to have been sent from a legitimate CDC email account – CDC-Covid19[@]cdc.gov. The emails include an attachment titled “Safety Precautions” which appears to be an Excel spreadsheet, but it actually a .exe executable file. Double clicking on the file attachment triggers the download of a banking Trojan.
Email and text-based phishing campaigns are targeting UK taxpayers and impersonate HM Revenue and Customs (HMRC). The emails include a legitimate HMRC logo and advise the recipients about a new COVID-19 tax refund program. According the emails, the refund program was set up in cooperation with National Insurance and National Health Services and allows taxpayers to claim back tax to help deal with the coronavirus pandemic. In order to receive the refund, the user is told they must supply their name, address, mother’s maiden name and their bank card number.
In the past few days, a web-based malware distribution campaign has been identified. Several websites are now displaying world maps and dashboards that allow people to track the spread of the virus and find out about the location of new cases. People are naturally concerned about cases in their local area, and the website maps are attracting a lot of visitors.
Shai Alfasi, a security researcher at Reason Labs, discovered several websites using fake versions of maps and dashboards. The websites prompt users to download an application that allows them to track infections in real-time. The application is an executable file that delivers the AZORult information stealer.
With COVID-19 infections increasing and showing no sign of slowing, COVID-19 phishing campaigns are likely to continue. Organizations should raise awareness of the threat of COVID-19 phishing attacks with their employees and ensure appropriate technical solutions are implemented to block web and email-based attacks. TitanHQ can help with the latter and can provide advanced email and web security solutions to block these attacks. If you have not yet implemented a web filter or email security solution to protect your Office 365 accounts, now is a good time to start. Contact TitanHQ today for further information.
by titanadmin | Mar 11, 2020 | Phishing & Email Spam, Spam News |
Microsoft has announced it has taken control of the U.S. infrastructure of the Necurs botnet and has taken steps to prevent the botnet operators from registering new domains and the rebuilding the Necurs infrastructure.
The Scale of the Necurs Botnet
The Necurs botnet first appeared in 2012 and has grown into one of the largest spam and malware distribution networks. The botnet consists of around 9 million devices that have been infected with Necurs malware. Each device within the botnet is under the control of the cybercrime group behind the botnet.
The Necurs botnet is used to commit a wide range of cybercrimes by the operators of the botnet as well as other cybercriminal groups who rent out parts of the botnet as a service. The Necurs botnet was used for malware and ransomware distribution, cryptocurrency mining, and attacks on other computers to steal credentials and confidential data. The Necurs botnet also has a distributed Denial of Service (DDoS) module capable of performing massive DDoS attacks, although this function is yet to be used.
The main use of the botnet is spamming. The botnet has been used to send vast quantities of spam email, including emails pushing fake pharmaceutical products, pump and dump stock scams, and Russian dating scams. To give an example of the scale of the spamming, over a 58-day period of observation, Microsoft found that a single Necurs malware-infected computer had sent out 3.8 million spam emails to 40.6 million email accounts. That is just one infected device out of 9 million! In 2017, the botnet was being used to spread Dridex and Locky ransomware at a rate of around 5 million emails an hour and between 2016 and 2019 the botnet was responsible for 90% of email-based malware attacks.
The Takedown of Necurs Infrastructure
Microsoft has tracked the criminal activity of the Necurs botnet operators for 8 years. The gang is believed to be Evil Corp, the Russian cybercriminal group behind the Dridex banking Trojan. Evil Corp has been named the most harmful cybercrime group in the world.
The takedown of the Necurs botnet involved a coordinated effort by Microsoft and partners in 35 countries. Microsoft obtained an order from the U.S. District Court for the Eastern District of New York on March 5, 2020 to seize the U.S. domains used by the botnet operators. These domains were used to issue commands to the 9 million infected computers.
Simply seizing the domains would not be sufficient to take down the botnet, as the botnet’s command and controls servers could be rapidly rebuilt. Domains used by the threat actors are often taken down, so new domains are constantly registered weeks or months in advance.
The key to long-term disruption of the botnet was cracking the algorithm used by the threat actors to generate new domains. Microsoft analyzed the algorithm and calculated more than 6 million domains that would be used by the threat actors over the next 25 months. Steps have been taken to prevent those domains from being registered and becoming part of the Necurs infrastructure.
The 9 million devices around the world are still infected with Necurs malware. Microsoft and its partners have identified the infected devices and are working with ISPs and CERT teams around the world to rid those devices of the malware.
by titanadmin | Feb 17, 2020 | Phishing & Email Spam, Spam News |
Just a few days after new figures from the FBI confirmed business email compromise scams were the biggest cause of losses to cybercrime, news broke of a massive cyberattack on a Puerto Rico government agency. Cybercriminals had gained access to the email account of an employee, understood to work in the Puerto Rico Employee Retirement System.
The compromised email account was used to send requests to other government agencies requesting changes be made to standard bank accounts for remittance payments. Since the email account used was trusted, the changes to bank accounts were made. Scheduled payments were then made as normal and millions of dollars of remittance payments were wired to attacker-controlled bank accounts.
The Puerto Rico Industrial Development Company, a state-owned corporation that drives economic development of the country, was one of the worst hit. Emails were received requesting changes to bank accounts and two payments were made. The first payment of $63,000 was made in December and another payment of $2.6 million in January. Other departments were also targeted, including the Tourism Company. The latter made a payment of $1.5 million. In total, the scammers attempted to steal around $4.73 million.
The business email compromise scam was uncovered when those payments were not received by the correct recipients. Prompt action was then taken to block the transfers and some of the payments were frozen, but the government has not been able to recover around $2.6 million of the stolen funds.
A full investigation has been launched to determine how the attackers gained access to the email account to pull off the scam. While the method used has not been confirmed, BEC attacks usually start with a spear phishing email.
A phishing email is sent to a person of interest requesting urgent action be taken to address a problem. A link is supplied in the email that directs the user to a website that requests their email account credentials. The account can then be accessed by the attacker. Attackers often set up mail forwarders to receive a copy of every email sent to and from the account. This enables them to learn about the company and typical payments and construct highly convincing scam emails.
Once access to a corporate email account is gained, the BEC scam is much harder to identify and block. The best defense is to ensure that the initial phishing emails are not delivered, and that is an area where TitanHQ can help.
by titanadmin | Feb 17, 2020 | Phishing & Email Spam |
A new report from the FBI’s Internet Crime Complaint Center (IC3) has revealed the extent to which phishing is used to attack businesses and the huge losses that have resulted from another form of email attack – business email compromise (BEC) scams.
In 2019, IC3 received 467,361 complaints about cybercrime and there were reported losses in excess of $3.5 billion, up from $2.7 billion in 2018. The true losses and number of attacks will be far higher, as not all crimes and losses are reported. Phishing, vishing, smishing, and pharming attacks were the most prevalent crime types with 114,702 complaints submitted to IC3 in 2019. Those attacks resulted in losses of more than $57 million.

Source: IC3
There were 23,775 complaints about BEC attacks and losses to those attacks were more than $1.776 billion. On average, BEC attacks result in losses of around $75,000 and the attacks accounted for 50.75% of all losses to cybercrime in 2019.
Business email compromise attacks involve the impersonation of a known individual or company and a fake invoice and fraudulent wire transfer request. Alternatively, changes to vendor’s bank account details or requested or changes to direct deposit accounts for payroll. These email impersonation attacks involve spoofing an email account or compromising an account, with the latter usually achieved with phishing emails.
Email is also used to deliver ransomware – 2,0417 incidents and $8,965,847 in losses – and malware and viruses – 2,373 incidents and $2,009,119 in losses.
The Importance of a Layered Approach to Email Security
As the IC3 2019 cost of cybercrime report shows, the most common attack vector is email, so how can business owners protect against email-based attacks?
Businesses can either purchase cybersecurity solutions directly or engage a managed service provider to look after cybersecurity. If the decision is taken to manage cybersecurity in-house, it is essential to adopt a defense in-depth strategy and implement multiple layers of protection. Should one cybersecurity solution fail to block a threat, other layers will prevent the attack from succeeding.
Many businesses have adopted Office 365 and use it for email. Microsoft includes a basic level of email protection for Office 365 as standard – Exchange Online Protection (EOP). EOP serves as the first layer of protection against phishing attacks, malware, and spam, but EOP alone is not enough to block sophisticated phishing attacks, BEC attacks, and zero-day malware threats. An additional layer of protection is required.
Advanced Protection Against Phishing and Business Email Compromise Attacks
TitanHQ has developed an advanced anti-spam solution – SpamTitan – that provides an additional layer of protection against email threats.
To protect against known malware threats, dual anti-virus engines are used. However, new malware variants are constantly being released. Before AV engines can block these new threats, the threat must be identified and the malware signature is then added to the AV engine’s virus definitions. Until that happens, threats will not be identified as malicious and will be delivered to inboxes.
To improve protection against zero-day threats, TitanHQ uses sandboxing fro email. When a suspicious or unknown email attachment is received, it is sent to the sandbox where it is subjected to in-depth analysis to identify command and control center callbacks and potentially malicious actions.
Office 365 accounts are targeted by cybercriminals and their new phishing campaigns are tested against Office 365 protections to make sure the emails are delivered. One previous study showed that 25% of phishing emails are delivered to Office 365 inboxes.
To ensure phishing threats are detected that would otherwise not be blocked by EOP, SpamTitan uses a range of advanced detection techniques. They include multiple real-time blackhole lists and threat intelligence feeds, multi-layered message analysis, SURBL’s, Bayesian analysis, greylisting, and more. Protection against email impersonation attacks and spoofing is provided through Sender Policy Framework and DMARC, and all outbound emails are scanned to identify potential email account compromises.
SpamTitan is a full-service email security solution that protects your business, your employees, and your clients from email-based attacks. With SpamTitan, you can adopt a layered approach to email security at a very low cost per user.
If you want to make sure that your business is protected from costly email-based attacks, give the TitanHQ team a call.
by titanadmin | Feb 12, 2020 | Email Scams, Phishing & Email Spam |
Emotet is the biggest malware threat faced by businesses and activity has increased considerably in recent weeks after a lull in December. Several new campaigns are now being identified each week, most of which are target businesses. One of the most recent campaigns uses a tried and tested technique to install the |Emotet Trojan. Malicious Word documents masquerading as invoices, estimates, renewals, and bank details.
The campaign mostly targets organizations in the United States and the United Kingdom, although attacks have also been detected in India, Spain, and the Philippines. Approximately 90% of emails in this campaign target financial services, with around 8% of attacks on companies in the food and drink industry.
The malicious Word documents are either attached to emails or hyperlinks are included in the emails that direct the user to a compromised website where the Word document is downloaded. The websites used are frequently changed and new Emotet variants are frequently released to prevent detection. Email security solutions that rely on AV engines to detect malware are unlikely to detect these zero-day threats as malicious.
Since Emotet is a massive botnet, emails spreading the Emotet Trojan come from many different sources. Email security solutions that rely on real-time blacklists are unlikely to detect these sources as malicious.
Emotet is primarily distributed via email from infected devices, but recently another distribution method has been identified. Emotet also spreads via Wi-Fi networks. This method has been used for almost two years, but it has only just been detected by security researchers at Binary Defense.
When Emotet is installed, a worm.exe binary is dropped that runs automatically. It attempts to connect to nearly Wi-Fi networks and brute forces weak passwords. Once connected to a Wi-Fi network, a search is conducted for non-hidden shares on the network. An attempt is made to enumerate all users connected to the Wi-Fi network, devices are brute forced, and the Emotet binary is dropped.
How to Block Emotet
The constantly changing tactics of the Emotet gang make detection difficult and no single solution will provide protection against all forms of attack. What is needed is a defense in-depth approach and layered defenses.
The primary defense against a predominantly email-based threat such as Emotet is an advanced spam filtering solution. Many businesses have used Office 365 and rely on the protection provided by Exchange Online Protection (EOP), which is included as standard with Office 365 licenses. However, EOP alone will not provide enough protection against Emotet. EOP will block all known malware threats, but it struggles to identify zero-day attacks. To block zero-day attacks, more advanced detection methods are required.
SpamTitan has been developed to work seamlessly with EOP to protect Office 365 email from zero-day threats. SpamTitan uses a variety of techniques to identify Emotet, including dual antivirus engines to block known Emotet variants and sandboxing to block zero-day attacks. Suspicious or unknown attachments are sent to the sandbox where they are subjected to in-depth analysis to identify command and control server callbacks and other malicious actions. SpamTitan also scans outgoing emails to identify attempts to spread Emotet from an already-infected machine. SpamTitan also incorporates DMARC to identify email impersonation and domain spoofing, which are commonly used in emails spreading Emotet.
To provide protection against the web-based element of attacks, including Emotet emails that use malicious hyperlinks rather than email attachments, another layer needs to be added to cybersecurity defenses – a DNS filtering solution such as WebTitan.
WebTitan uses real-time URL threat detection powered by 650 million end users. The real-time database includes more than 3 million malicious URLs and IP addresses and each day around 100,000 new malicious URLs are detected and blocked. WebTitan also includes real-time categorization and detection of malicious domains, full-path URLs, and IPs, with up-to-the-minute updates performed to block new malicious sources. As soon as a URL is identified as being used to distribute Emotet (or other malware) it is blocked by WebTitan. WebTitan also conducts link & content analysis, static, heuristic, & behavior anomaly analysis, and features in-house and 3rd party tools and feeds to keep users protected from web-based threats.
Other essential steps to take to tackle the threat from Emotet include:
- Disable macros across the organization
- Ensure operating systems are kept up to date and vulnerabilities are promptly patched.
- Set strong passwords to thwart brute force attacks
- Ensure endpoint protection solutions are deployed on all devices
- Provide security awareness training to employees
- Conduct phishing simulation exercises to identify employees that require further training
by titanadmin | Feb 10, 2020 | Internet Security, Network Security |
The first California Consumer Privacy Act lawsuit has been filed over an alleged failure to adequately protect consumer data. The lawsuit has been filed against Hanna Andersson, a children’s clothing company, and its ecommerce platform provider, Salesforce.com.
The California Consumer Privacy Act took effect on January 1, 2020. Under Civil Code 1798.100 – 1798.199, consumers could start exercising their new rights under CCPA from the compliance date. One of those rights is being able to take legal action against companies for privacy violations, such as the theft of personal data in a data breach.
The California Consumer Privacy Act lawsuit was filed in the U.S. District Court for the Northern District of California on behalf of a victim of a 2019 data breach. The lawsuit alleges negligence and a failure to implement reasonable safeguards to protect consumer data, and that the data breach occurred as a direct result of the alleged negligence. A claim for damages has not been stated, although the right has been reserved to seek damages and relief at a later date.
The breach in question was announced by Hanna Andersson on January 15, 2020. Hackers had gained access to its systems and downloaded malware, which allowed the attackers to steal information such as names, personal information, and payment card data. That information was subsequently listed for sale on the dark web.
The California Consumer Privacy Act allows Californians to file for damages of up to $750 per data breach, so a class action California Consumer Privacy Act lawsuit arising from a sizeable data breach could prove extremely costly for a company. In this case, the data breach affected approximately 10,000 California residents, so damages up to $7,500,000 could potentially be claimed.
Enforcement of CCPA
Enforcement of compliance by the California Attorney General has been delayed and will start 6 months after the publication of the final regulations or July 1, 2020, whichever comes sooner. Since the final regulations have yet to be published, the enforcement date will be July 1, 2020. California Attorney General Xavier Bercerra has already stated that he will make an example of businesses that fail to comply with CCPA.
It should be noted that there is nothing in CCPA that prevents the state attorney general from issuing notices of noncompliance before that date and consumers can already file lawsuits to claim damages. It is therefore essential for all entities covered by CCPA to ensure that they are honoring the new consumer rights and have implemented safeguards to protect consumer data.
How TitanHQ Can Help with CCPA Compliance
TitanHQ offers two powerful security solutions that can help covered entities ensure the data of consumers is protected and data breaches are prevented. These two cybersecurity solutions protect against the two most common attack vectors – Email and the internet.
SpamTitan is a powerful anti-spam, anti-malware, and anti-phishing solution that protects email systems from phishing and spear phishing attacks, known and zero-day malware threats, and email-based ransomware attacks.
WebTitan is a companion solution that blocks the web-based element of phishing attacks, exploit kits, and drive-by malware downloads over the internet, while also controlling the content that employees can access on wired and wireless networks.
TitanHQ can also help covered entities comply with the right to know and right to delete consumer rights afforded by CCPA through ArcTitan. ArcTitan is an email archiving solution that allows organizations to meet state and federal email data retention requirements and quickly find emails containing consumer data. If a California resident exercises their right to know what data is held on them by a company, or requests all of their personal data is deleted, that information can quickly be found in the archive. ArcTitan will also allow you to quickly find email data for eDiscovery in the event of any legal disputes.
For further information on these solutions, to schedule a product demonstration, or to arrange a free trial of the full solutions (with full customer support), give the TitanHQ team a call today.
by titanadmin | Feb 5, 2020 | Email Scams, Phishing & Email Spam, Spam Software |
Tax season is now underway and business email compromise scammers have stepped up their efforts to obtain W-2 forms for tax fraud. These attacks often start with spear phishing emails targeting the CEO and the executive board. Once email credentials have been obtained, the accounts are then accessed, and emails are sent internally to payroll and the HR department requesting the W-2 forms of employees who have worked in the previous tax year.
Scammers target businesses as there is much greater potential for profit than attacks on individual taxpayers, although consumers also need to be wary of IRS-related phishing scams. This time of year sees an increase in IRS phishing scams. Scammers impersonate the IRS and send emails informing taxpayers about a tax refund that is due and demands are sent for outstanding tax, with threats of dire consequences if prompt action is not taken to address issues.
Advances in email security have meant cybercriminals have had to get creative as it is harder to sneak phishing emails past email defenses. Phishing scams are now commonly initiated via text message, post, and over the telephone. There has already been one campaign identified where consumers are being targeted using robocalls warning that Social Security numbers have been suspended after suspicious activity was detected.
While many of these scams seek personal information, others are conducted to spread malware. One threat group that started its tax-related scams early this year is the Emotet gang. A campaign is currently being conducted that uses emails containing fake signed W-9 forms.
Signed W-9 forms are requested by companies from their contractors if they have been paid in excess of $600 during the tax year. Many companies will have requested signed W-9 forms from their contractors to confirm addresses and tax identification numbers, so they will be expecting copies of these forms in their inboxes.
The Emotet emails are short and to the point, saying “Thank you for your help. Pleased see attached file.” The emails include a Word document attachment named W-9.doc. When the document is opened, the Office 365 logo is displayed along with text stating the document was created in OpenOffice and requires the user to enable editing and enable content. Doing so triggers the silent download of the Emotet Trojan.
This is just one of the tax-related messages being used by the Emotet gang. There are likely to be many more variants sent over the next few weeks. Other cybercriminal gangs will similarly be conducting their own tax-themed phishing campaigns to spread different malware variants and ransomware.
Businesses, tax preparers, and consumers need to be on high alert during tax season for phishing scams and emails spreading malware.
Now is a good time for businesses to review their cybersecurity defenses and enhance protection against phishing and malware attacks. If you use Office 365 and rely on the anti-phishing protections built into Office 365 (EOP), you should consider enhancing your anti-phishing and anti-malware protection with a third-party spam filter – One that has superior malspam detection capabilities.
This is an area where TitanHQ can help. SpamTitan uses a variety of advanced techniques to detect and block phishing threats and zero-day malware, including an email sandbox where unknown and suspicious email attachments are subject to in-depth analysis. Give the TitanHQ team a call to find out more about SpamTitan, improving Office 365 malware and phishing protection, and to arrange a product demonstration and free trial of SpamTitan.
In the meantime, take steps to alert your workforce about tax-season phishing scams and prepare them in case a phishing email arrives in their inbox. An email alert sent to your employees about the threat of tax-season scams could prevent a costly phishing attack or malware infection.
by titanadmin | Jan 31, 2020 | Email Scams, Phishing & Email Spam, Spam News |
A novel coronavirus phishing campaign has been detected that uses scare tactics to trick users into infecting their computer with malware.
The World Health Organization has now declared the 2019 novel coronavirus outbreak a global emergency. The number of cases has increased 10-fold in the past week with almost 9,100 cases confirmed in China and 130 elsewhere around the world.
A worldwide health crisis such as this has naturally seen huge coverage in the press, so it is no surprise that cybercriminals are capitalizing on the concern and are using it as a lure in a malspam campaign to scare people into opening an email attachment and enabling the content.
A novel coronavirus phishing campaign has been detected that uses a fake report about the coronavirus to get email recipients to open a document that details steps that should be taken to prevent infection. Ironically, taking the actions detailed in the email will actually guarantee infection with a virus of a different type: Emotet.
The coronavirus phishing campaign was identified by IBM X-Force researchers. The campaign is targeted on users in in different Japanese prefectures and warning of an increase in the number of local confirmed coronavirus cases. The emails include a Word document attachment containing the notification along with preventative measures that need to be taken.
If the attachment is opened, users are told they must enable content to read the document. Enabling the content will start the infection process that will see the Emotet Trojan downloaded. Emotet is also a downloader of other malware variants. Other banking Trojans and ransomware may also be downloaded. Emotet can also send copies of itself to the victim’s contacts. Those messages may also be coronavirus related.
To add credibility, the Emotet gang makes the emails appear to have been sent by a disability welfare service provider in Japan. Some of the captured messages include the correct address in the footer.
More than 2,000 new infections have been confirmed in the past 24 hours in China and all of its provinces have now been impacted. Cases have now been reported in 18 other countries with Thailand and Japan the worst hit outside of China with 14 cases confirmed in each country. As the coronavirus spreads further and more cases are reported, it is likely that the Emotet gang will expand this campaign and start targeting different countries using emails in different languages. Kaspersky lab has also said that it has identified malspam campaigns with coronavirus themes that use a variety of email attachments to install malware.
Businesses can protect against Emotet, one of the most dangerous malware variants currently in use, by implementing a spam filtering solution such as SpamTitan that incorporates a sandbox where malicious documents can be analyzed in safety to check for malicious actions.
For further information on protecting your email system, contact TitanHQ today.
by titanadmin | Jan 30, 2020 | Industry News, Phishing & Email Spam, Spam News |
It has been well documented how much time businesses waste dealing with spam and there is no denying the threat that malicious spam emails (malspam) pose, but it is not just a problem for big business. Spam in academia is also a major problem.
A recent study published in the journal, Scientometrics, explores the cost of spam in academia. The study was primarily focused on spam emails sent by new, non-peer-reviewed journals that are attempting to gain a share of the market. These journals are adopting the same spam tactics often used by scammers to sell cheap watches and cut-price medications and for phishing and spreading malware.
Three researchers – Jaime A. Teixeira da Silva, Aceil Al-Khatib, and Panagiotis Tsigaria – attempted to quantify the amount of time that is being wasted dealing with those messages and the losses that result.
To assess the extent of the problem, the researchers used figures from several studies on spamming to obtain an average number of targeted spam emails that academics receive each day. They opted for a conservative figure of 4-5 messages, per academic, per day. Most of those messages take just a few seconds to open and read but that time mounts up. They assumed an average time of 5 seconds per message – less than half a minute per day. That equates to $100 per researcher, per year at an average hourly rate of $50. Using the United Nations estimate of the number of researchers in academia globally, the total global cost of spam in academia was estimated to be $1.1 billion a year.
That figure is based on the lost time alone and does not factor in non-targeted spam emails – bulk unsolicited emails not specifically targeting researchers. Add in the time dealing with those messages and the global cost reaches $2.6 billion a year. To put the cost into perspective, $2.6 million is much more than the time researchers devote to peer review, which has been estimated at a cost of $1.9 billion a year. The figures do not include the considerable losses due to phishing, malware, and ransomware attacks. Factor in those costs and the losses would be several orders of magnitude higher.
Co-author of the study, Panagiotis Tsigaris, a professor of economics at Thompson Rivers University in Canada, explained that there is no silver bullet when it comes to dealing with spam and suggested several ways that the cost of spam in academia could be reduced.
Tsigaris suggests that penalties should be increased for publishing in predatory journals, and that academics should be educated about spam email and that improvements should be made to email filtering technology.
Here at TitanHQ, we are well aware of the problem of spam, both in terms of the productivity losses that spam causes, and harm caused by malicious spam emails.
To help prevent losses and downtime due to spam and email-based threats, TitanHQ has developed a powerful, easy-to-use, and cost-effective cloud-based spam filtering solution called SpamTitan. SpamTitan has been independently tested and shown to block in excess of 99.9% of spam email, 100% of known malware and ransomware threats, and thanks to a host of detection measures and sandboxing, SpamTitan is also effective at blocking zero-day (new) malware and ransomware threats.
To find out more about SpamTitan and how you can block more spam and ensure malicious emails do not reach your researchers’ inboxes, give the TitanHQ team a call today.
by titanadmin | Jan 22, 2020 | Industry News, Internet Security, Network Security, Website Filtering |
TitanHQ has announced a new partnership with Pax8. The partnership means Pax8 partners now have access to TitanHQ’s cloud-based email security solution – SpamTitan – and its DNS filtering solution, WebTitan.
Pax8 is the leader in cloud distribution. The company simplifies the cloud buying process and empowers businesses to achieve more with the cloud. The company has been named Best in Show for two consecutive years at the Next Gen and XChange conferences and is positioned at number 60 in the 2019 Inc. 5000 list of the fastest growing companies.
Pax8 carefully selects the vendors it works with and only offers market-leading channel friendly solutions to its partners. When searching for further cybersecurity solutions for its partners, TitanHQ was determined to be the perfect fit. TitanHQ is the leading provider of cloud-based email and web security solutions for managed service providers (MSPs) serving the SMB marketplace and its cybersecurity solutions are much loved by users. This was clearly shown in the 2019 G2 Crowd Report on Email Security Gateways where SpamTitan was named leader, having achieved 4- or 5-star ratings by 97% of its users, with 92% saying they would recommend the solution to other businesses.
Phishing, malware, and ransomware attacks have all increased in the past year and the cost of mitigating those attacks continues to rise. By implementing SpamTitan and WebTitan, SMBs and MSPs can secure their email environments and block web-based threats and keep their networks secure.
SpamTitan provides excellent protection for Office 365 environments. The solution detects and blocks phishing and email impersonation attacks and prevents known and zero-day malware and ransomware threats from reaching inboxes. The WebTitan Cloud DNS filtering solution blocks the web-based component of cyberattacks by preventing end users from visiting malicious websites, such as those harboring malware and phishing kits.
Both solutions are quick and easy to implement, can be seamlessly integrated into MSPs service stacks and cloud-management platforms, and Pax8 partners benefit from highly competitive and transparent pricing, centralized billing, and leading customer support.
“I am delighted to partner with the Pax8 team,” said Ronan Kavanagh, CEO, TitanHQ. “Their focus and dedication to the MSP community are completely aligned with ours at TitanHQ, and we look forward to delivering our integrated solutions to their partners and customers.”
by titanadmin | Jan 21, 2020 | Email Scams, Phishing & Email Spam |
The Emotet botnet took a Christmas holiday but it’s now up and running again and the massive phishing and spamming campaigns have resumed. These campaigns, which involve millions of spam emails, use a variety of lures to trick people into opening an attachment and enabling content. The content in question includes a macro that runs a PowerShell command that downloads and executes the Emotet Trojan.
The Emotet Trojan is bad news. Emotet was once just a banking Trojan whose purpose was to steal online banking credentials. It still does that and much more besides. Emotet also steals credentials from installed applications and browsers. It is also self-propagating and will send copies of itself via email to the victim’s contacts. As if that was not bad enough, Emotet has another trick up its sleeve. It is also a downloader of other malware variants such as the TrickBot Trojan and Ryuk ransomware. These additional payloads allow data to be stolen and sold for profit and for files across the network to be encrypted and ransom demands issued. Emotet has also delivered cryptocurrency miners in the past and could deliver any number of other malware payloads.
The scale of the botnet is staggering. In the first quarter of 2019, Emotet was responsible for 6 out of 10 malicious payloads delivered via email. There are often breaks in activity, but even though the threat actors behind the botnet took almost half of 2019 off, Emotet still ranks as the top malware threat of the year.
Emotet sprung back to life on January 13, 2020, with targeted attacks on the pharmaceutical industry in North America, but it didn’t take long for the attacks to spread even further afield. Now more than 80 countries are being attacked and in addition to English, campaigns have been detected in Italian, Polish, German, Spanish, Japanese, and Chinese.
The lures used to fool end users into opening email attachments are highly varied and often change. Tried and tested lures such as fake invoices, orders, statements, agreements, payment remittance notices, receipts, and delivery notifications are often used in attacks on businesses, which are the primary targets. Before the botnet shut down for a break in December, Greta Thunberg-themed emails were being used along with Christmas party invitations. A host of new lures can be expected in 2020.
The themes of the emails may change but the messages have one thing in common. They require an end user to take action. That is usually opening a document, spreadsheet, or other file, but could be a click on a hyperlink in an email. Once that action is taken, Emotet will be silently downloaded.
There are two main ways of blocking attacks and both are necessary. The first is to ensure that the email system is secure, which means implementing an effective spam filter. Businesses that use Office 365 will have a modicum of protection through Exchange Online Protection (EOP), which is included with Office 365 subscriptions. However, businesses should not rely on EOP alone. Layered defenses are required.
SpamTitan is a powerful spam filter that will improve protection against malware threats such as Emotet. SpamTitan can be layered on top of Office 365 to provide greater protection and prevent the malware from being delivered to inboxes. Dual anti-virus engines are incorporated into the solution to detect known threats and SpamTitan includes a sandbox for identifying threats that signature-based detection mechanisms miss.
Many businesses deploy a variety of security solutions but fail to prepare their employees for an attack. If malicious emails make it past security solutions and are delivered to inboxes, all it takes is for one employee to fail to spot the threat and respond for Emotet to be installed (and potentially ransomware as well). It is therefore important to provide regular security awareness training to everyone in the company from the CEO down. If employees are not told how to identify malicious emails, they cannot be expected to spot threats and report the messages to the security team.
Fortunately, through a combination of email security solutions and security awareness training, the threat from Emotet can be neutralized. For more information on the former, give TitanHQ a call today.
by titanadmin | Jan 20, 2020 | Email Scams, Phishing & Email Spam |
Whenever there is a major event that attracts a lot of media attention cybercriminals will be poised to take advantage, so it is no surprise that warnings are being issued about Travelex phishing scams.
The Travelex ransomware attack that struck on New Year’s Eve involved a ransomware variant called Sodinokibi. The gang responsible is one of the most prolific threat groups using ransomware. The group’s attacks are highly targeted and seek to encrypt entire networks and the ransom demands reflect the scale of encryption. Travelex was initially issued with a demand for a payment of $3 million. That soon doubled to $6 million when payment was not made within the allocated timescale.
The fallout from the attack has been immense, which is unsurprising given that Travelex is the largest provider of currency exchange services worldwide. Many banks and retailers rely on Travelex to provide for their currency exchange services. Without access to those online services, currency exchange services came to a grinding halt. It has taken two weeks for Travelex to start bringing some of its services back online, but its website remains down and the disruption continues.
The attackers claimed to have stolen large quantities of customer data from Travelex. The attackers threatened to publish or sell the data if the ransom was not paid. This tactic is becoming increasingly common with ransomware gangs. In this case, the sodinokibi gang claimed to have gained access to Travelex systems 6 months previously and said they had stolen customer data including names, payment card information, and Social Security numbers and National Insurance numbers. The gang had also recently attacked the American IT company Artech Systems and had posted 337MB of data stolen in that attack, demonstrating to others that it was not an empty threat. Travelex maintained that no customer data had been stolen, but that has yet to be confirmed.
Warning Issued About Travelex Phishing Scams
Travelex customers should naturally err on the side of caution and monitor their accounts for signs of fraudulent use of their information but there are other risks from an attack such as this.
Travelex has issued a warning to its customers recommending they should be alert to the threat of phishing attacks via email and over the phone. Opportunistic scammers often take advantage of major events such as this and Travelex phishing scams are to be expected, as was the case following the TalkTalk data breach. These phishing scams are likely to be most effective on Travelex customers who have lost money as a result of the attack. Any offer of compensation or a refund is likely to attract a response.
For consumers, the advice is never to open email attachments or click on links in unsolicited emails. Businesses should also take steps to protect their networks from malware and phishing attacks.
Businesses should adopt a defense in depth strategy to protect against phishing scams and malware attacks. An advanced email security solution such as SpamTitan should be used to protect Office 365 accounts. SpamTitan improves protection against zero-day malware and phishing threats and blocks threats at the gateway.
A web filtering solution such as WebTitan should be used to block the web-based component of phishing and malspam campaigns and prevent end users from visiting malicious websites. End user training is also a must. It is important to teach employees how to identify phishing emails and malspam, and condition them how to respond when suspicious emails are received.
by titanadmin | Jan 16, 2020 | Email Scams, Phishing & Email Spam |
A new ransomware threat – Ako ransomware – has emerged which is targeting business networks and is being distributed via spam email. The ransomware is being offered to affiliates under the ransomware-as-a-service model and the aim of the attackers is clear. To maximize the probability of payment of the ransom by making recovery harder, and to steal data prior to encryption to ensure the attack is still profitable if the ransom is not paid. Having the data could also help convince the victims to pay up, as we have seen in recent attacks involving Maze and Sodinokibi ransomware, where threats are issued to publish stolen data if the ransom is not paid.
The developers of Ako ransomware appear to be going for large ransom payments, as they are not targeting individual workstations, rather the entire network. The ransomware scans local networks for other devices and will encrypt network shares. The ransomware deletes shadow copies and recent backups and disables Windows recovery to make recovery more difficult without paying the ransom.
Encrypted files are given a randomly generated file extension and retain the original file name. No ransom amount is stated in the ransom note. Victims are required to contact the attackers to find out how much they will need to pay for the keys to decrypt their files.
One of the intercepted emails being used to distribute the ransomware uses a password-protected zip file as an attachment. The email appears to be a business agreement which the recipient is asked to check. The password to open and extract the file is included in the message body. The zip file attachment – named agreement.zip – contains an executable file which will install Ako ransomware if it is run. The malicious file is called agreement.scr.
There is no free decryptor for Ako ransomware. Recovery without paying the ransom will depend on whether viable backups exist that have not also been encrypted. It is therefore important to make sure backups are regularly performed and at least one copy of the backup is stored on a non-networked device to prevent it also being encrypted by the ransomware. Backups should also be tested to make sure file recovery is possible.
Since Ako ransomware is being distributed via spam email, this gives businesses an opportunity to block an attack. An advanced spam filtering solution should be implemented that scans all inbound messages using a variety of detection mechanisms to identify malware and ransomware threats. A sandbox is an important feature as this will allow email attachments to be analyzed for malicious activity. This feature will improve detection rates of zero-day threats.
nd user training is important to ensure that employees do not open potentially malicious files. Training should condition employees never to open email attachments in unsolicited emails from unknown senders. As this campaign shows, any password protected file sent in an unsolicited email is a big red flag. This is a common way that ransomware and malware is delivered to avoid detection by antivirus solutions and spam filters.
Anti-spam solutions and antivirus software will not be able to detect the threat directly if malicious files are sent in password-protected archives, which can only be opened if the password is entered. Rules should therefore be set to quarantine password-protected files, which should only be released after they have been manually checked by an administrator. With SpamTitan, these rules are easy to set.
Ako ransomware is one of many new ransomware threats that have been released in recent months. High profile attacks on companies such as Travelex that see massive ransom demands issued, which in many cases are paid, show a huge payday is possible.
Ransomware developers will keep developing new threats for as long as attacks remain profitable, and there is not likely to be a shortage of affiliates willing to run spamming campaigns to get their slice of the ransom payments.
With the attacks increasing, it is essential for you to have strong defenses that can detect and block malware, ransomware, and phishing threats, and that is an area where TitanHQ can help.
To find out more about how you can improve your defenses against email and web-based threats, give the TitanHQ team a call today.
by titanadmin | Jan 15, 2020 | Network Security |
The Travelex ransomware attack that started around December 31, 2019 is one of several recent ransomware attacks where threat actors have upped the ante by threatening to publish data stolen from victims prior to the deployment of ransomware.
A New Trend in Ransomware Attacks
Most ransomware attacks, especially those conducted by affiliates using ransomware-as-a-service, see ransomware deployed instantly. An employee receives a ransomware attachment via email, opens the attachment, and the encryption process is started. Now, several threat actors have taken steps to increase the probability of their ransom demand being paid.
The U.S. Department of Homeland Security’s Cybersecurity and Infrastructure Security Agency has recently issued warnings about changing ransomware tactics, which now involve data theft prior to file encryption. This tactic is nothing new, as several threat actors have been conducting these types of attacks for some time, attacks of this nature have been increasing.
to the network is gained, the attackers then move laterally and gain access to as many devices as possible. Data is stolen and when the attackers have stolen as much as they want, ransomware is deployed. In these types of attacks, the time between the initial compromise and deployment of ransomware is typically several months.
Data may be stolen and sold online with the ransomware deployed as a coup de grace after a long-term compromise to extort money from the company. Now it is increasingly common for a threat to be issued along with the ransom demand that the stolen data will be published or sold if the ransom is not paid.
This tactic has been adopted by the threat actors behind Maze ransomware and they have gone ahead and published stolen data when the ransom was not paid. The threat actors using MegaCortex ransomware and LockerGoga ransomware have similarly issued threats.
Now the gang behind Sodinikibi (REvil) ransomware have also changed tactics and have started issuing threats to publish stolen data. The Sodinokibi gang have made several threats to sell on or publish stolen data but it was only recently that they did just that. The gang attacked Artech Information Systems, one of the largest IT staffing companies in the U.S. When the ransom demand was not paid, 337MB of stolen data was published on a Russian hacking and malware forum. The Travelex ransomware attack is one of the latest Sodinokibi ransomware attacks, and a threat to publish stolen data was similarly issued.
The Travelex Ransomware Attack
On New Year’s Eve, Travelex took its systems offline to contain the infection and limit the damage caused. More than two weeks on, Travelex systems are still offline although the company is now starting to restore some of its systems. The number of branches affected by the attack, and banks and other companies that rely on its currency exchange services, makes this one of the most serious and damaging ransomware attacks ever.
With its systems offline, Travelex has been unable to provide its currency services to banks such as HSBC, Royal Bank of Scotland, NatWest, First Direct, Barclays and Lloyds, all of which rely on Travelex for providing their currency services. Many other companies, such as the supermarket chains Sainsbury’s and Tesco, have also had to stop providing online currency services to their customers. Travelex has been forced to provide services manually using pen and paper for over the counter currency exchanges in its branches. More than 70 countries in which Travelex operates were affected by the attack.
Travelex has only released a limited amount of information about the attack, but the attackers have been in contact with several media outlets. Initial reports suggested a payment of $3 million was required for the keys to unlock the encryption, although the demand doubled to $6 million when payment was not received within the stipulated 2 days. The attackers also threatened to publish data stolen in the attack if the payment was not made within 7 days.
Travelex issued a statement saying no customer data was breached and that the infection was contained, a position that has been maintained since the attack, even though the Sodinokibi gang has threatened to publish customer data.
The Sodinokibi ransomware gang, through a spokesperson, said the gang had stolen 5GB of customer data including customers’ names, dates of birth, credit card information, Social Security numbers, and National Insurance numbers. The gang claimed that all stolen data would be deleted and would not be used if the ransom demand was paid, but that the data would be sold if payment was not received. The gang also said access to Travelex systems was gained 6 months before the ransomware was deployed.
How Was Travelex Attacked?
It is not known at this stage exactly how ransomware was installed on its network, but there have been several security researchers that have offered some clues. According to BleepingComputer, Travelex was using insecure services prior to the attack. Security researcher Kevin Beaumont found Travelex had AWS Windows servers that did not have Network Level Authentication enabled, which could have given the attackers the opportunity they needed to launch an attack.
A critical vulnerability in the Pulse Secure VPN enterprise solution for secure communications – CVE-2019-11510 – was identified and was patched by Pulse Secure on April 24, 2019, but many companies were slow to apply the patch, despite receiving multiple warnings from Pulse Secure. An exploit for the vulnerability was made public on August 21, 2019.
Troy Mursch, chief research officer at Bad Packets, found that Travelex had not applied the patch by the time the exploit was released. The Sodinokibi ransomware gang said they compromised Travelex 6 months prior to the deployment of ransomware. This could have been the vulnerability that was exploited.
Recovery Now Well Underway
On January 13, 2020, more than 2 weeks after the ransomware attack was experienced, Travelex issued a statement confirming that the recovery process was well underway, although the firm’s website was still offline. The company had started restoring its currency services to banks and its own network. Internal order processing has been restored and customer-facing systems are slowly being brought back online. What Travelex has not confirmed is whether the ransom was paid. No Travelex data appears to have been published online so it is possible that a ransom payment has been negotiated with the attackers.
Cost of the Travelex Ransomware Attack
The ransom payment is considerable but is likely to be several orders of magnitude less than the costs of downtime and disruption to its services.
No customer data appears to have been misused, but Travelex could still face a barrage of lawsuits from customers and the Information Commissioner’s Office and other data protection authorities my choose to fine Travelex over the data breach, either for the exposure of data or for the failure to report under GDPR.
GDPR requires data breaches to be reported to data protection authorities within 72 hours and it appears that did not happen. The maximum financial penalty for a GDPR violation is €20 million or 4% of a company’s global annual turnover, whichever is greater. Travelex’s global annual turnover in 2018 was $947.86 million. A fine of $189.57 million could therefore be issued. It should be noted that even if data was not stolen by the attackers and was just made inaccessible, it still counts as a reportable data breach under GDPR.
A payment of $6 million to the attackers would only be a tiny proportion of the total losses from downtime, lost business, lawsuits, and regulatory fines.
by titanadmin | Dec 30, 2019 | Email Scams, Phishing & Email Spam |
Customers of Canadian banks have been targeted by cybercriminals in an extensive phishing campaign that has been ongoing for at least the past two years, according to Check Point Research which uncovered the campaign. As with many other financial phishing scams, the attackers spoof the website of a well-known bank and create a virtual carbon copy of the home page of the bank on a lookalike domain, which often only differs from the genuine domain name by a letter or two.
A link to the fraudulent site is then sent in a mass spamming campaign to email addresses on the specific country top level domain where the bank operates. The emails instruct users to visit the banks website and login, usually under the guise of a security alert. When the link in the email is clicked, the user is directed to the spoofed site and may not notice the domain name is not quite right. They then enter their login credentials which are captured by the scammers. The credentials are then used to make fraudulent wire transfers to accounts controlled by the attackers.
In this campaign, the emails include a PDF email attachment. PDF files tend to be trusted to a higher degree than Word documents and spreadsheets, which end users have usually been instructed to treat as suspicious. The PDF file includes a hyperlink, which the user is instructed to click. Since the hyperlink is in the document rather than the email body, it is less likely to be scanned by email security solutions and has a higher chance of being delivered.
The user is told that they are required to update their digital certificate to continue using the online banking service. The PDF file includes the bank logo and a security code, which the user is required to enter when logging in. The code is included in the PDF attachment rather than email body for security reasons. As with most phishing scams, there is urgency. The recipient is told that the code expires in 2 days and that they must register within that time frame to avoid being locked out of their account.
The landing pages on the websites are identical to those used by the banks as the attackers have simply taken a screen shot of the bank’s landing page. Text boxes have been added where the username, password, and token number must be entered. Users are then asked to confirm the details they entered while the attackers attempt to access their account in real-time and make a fraudulent transfer.
These tactics are nothing new. Scams such as this are commonplace. What is surprising is how long the campaign has been running undetected. The scammers have been able to operate undetected by registering many lookalike domains which are used for a short period of time. Hundreds of different domains have been registered and used in the scam. At least 14 leading banks in Canada have had their login pages spoofed including TD Canada Trust, Scotiabank, Royal Bank of Canada, and BMO Bank of Montreal.
All of the websites used in the scam have now been taken down, but it is all but guaranteed that other lookalike domains will be registered and further scams will be conducted.
by titanadmin | Dec 27, 2019 | Email Scams, Phishing & Email Spam |
A spamming campaign has been detected that is piggybacking on the popularity of Greta Thunberg and is using the climate change activist’s name to trick individuals into installing the Emotet Banking Trojan.
Emotet is one of the most active malware threats. Emotet was first detected in 2014 and was initially used to steal online banking credentials from Windows users by intercepting internet traffic. Over the years it has undergone several updates to add new functionality. It has had a malspam module added, which allows it to send copies of itself via email to a user’s contacts. Emotet also includes a malware downloader, allowing it to download a range of other malware variants such as other banking Trojans and ransomware.
The malware is used indiscriminately in attacks on individuals, businesses, and government agencies, with the latter two being the main targets. Emotet is primarily spread via spam email, and while exploits are not used to spread to other devices on the network – EternalBlue for instance – other malware variants downloaded by Emotet can. TrickBot for instance.
The Greta Thunberg spam campaign aims to get users to open a malicious Word attachment and enable content. If that happens, Emotet will be silently downloaded to the user’s device, sensitive banking information will be stolen, and further malware may be downloaded.
The campaign was active over the holiday period and used a variety of Christmas-themed lures to entice users into opening the email attachment. Some of the emails did not include an attachment and instead used a hyperlink to direct the user to a website where the malicious document could be downloaded.
One of the emails wished the recipient a Merry Christmas and urged them to consider the environment this Christmastime and join a demonstration in protest against the lack of action by governments to tackle the climate crisis. The email claimed details about the time and location of the protest were included in the Word document. The email also requested the recipient to send the email on to all their colleagues, friends, and relatives immediately to get their support as well. Several variations along that theme have been detected.
To increase the likelihood of the recipient enabling content, when opened the document displays a warning that appears to have been generated by Microsoft Office. The user is told that the document was created in OpenOffice and it is necessary to first enable editing first and then enable content. Doing the latter will enable macros which will start the infection process.
The emails are well written and have been crafted to get an emotional response, which increases the likelihood of the user taking the requested action. The emails have been sent in multiple languages in many different countries.
Whenever there is a major news event, popular sports tournament, or other event that attracts global interest, there will be cybercriminals taking advantage. Regardless of the theme of any email, if it is unsolicited and asks you to click a link or open an email attachment, it is best to assume that it is malicious.
Businesses can protect their networks against threats such as these by implementing an advanced spam filtering solution such as SpamTitan. SpamTitan will identify threats such as phishing attacks and will prevent the messages from reaching inboxes. SpamTitan also includes dual anti-virus engines to detect known malware and machine learning techniques and sandboxing to identify and block zero-day malware.
For further information on how SpamTitan can protect your business from email threats such as this, contact TitanHQ today.
by titanadmin | Dec 23, 2019 | Spam Advice, Spam Software |
The majority of businesses have experienced a phishing attack in the past year, and according to one survey on SMBs in the United States, 72% have experienced a phishing attack in the past 3 months.
In healthcare, phishing is the leading cause of data breaches by some distance. In November 2019, there were 17 phishing-related data breaches reported to the Department of Health and Human Services Office for Civil Rights out of 33 for the month. Since OCR only makes breach reports public if they have resulted in the exposure of 500 or more records, the total number of phishing attacks is likely to be substantially higher.
Phishing attacks are increasing, and the reason is simple. Phishing is the easiest way of attacking an organization to deliver malware or obtain sensitive information. That is because phishing targets the weakest link: Employees. Employees are getting better at identifying phishing emails through security awareness training, but cybercriminals have responded and are now conducting highly sophisticated phishing attacks that are much harder for employees to identify.
There has also been an increase in spear phishing attacks. This is a much more targeted form of phishing. Instead of millions of emails being sent out in a campaign, only a handful are sent or to very specific targets. The emails are written to maximize the chances of success and are usually personalized.
So how can a business improve its defenses against phishing and spear phishing? Unfortunately, there is no silver bullet. Businesses need to take a defense in depth approach to significantly improve resilience to phishing attacks.
The best place to start is with an advanced email security solution. Phishing requires some form of manual action in order to succeed. If you prevent phishing emails from reaching inboxes, employees will not be able to click on links or download malware. An advanced email security solution will be able to block the vast majority of phishing emails before they reach your email system.
You will no doubt already have a spam filtering solution in place, but is it effective? Are phishing emails still being delivered? One common mistake made by SMBs is to believe that their Office 365 environment is well protected by default, when the reality is Exchange Online Protection (EOP) that comes with Office 365 fails to block many phishing attempts. One study showed 25% of phishing emails were not blocked by EOP. If you want to improve your defenses against phishing, you should use a third-party anti-spam and anti-phishing solution on top of EOP: One that compliments EOP but provides greater protection. SpamTitan for example.
With more phishing emails being blocked, your security posture will be much improved, but you can’t stop there. No anti-phishing solution will block all phishing threats, 100% of the time. Since all it takes is for one phishing email to be clicked for a data breach to occur, you need to add another layer to your defenses.
A DNS filtering solution provides protection against the web-based part of phishing attacks. When an employee clicks a link in an email and is directed to a fake Office 365 login page or a site where malware is downloaded, the attempt to access the site will be blocked.
A DNS filter blocks attempts to access phishing sites at the DNS lookup stage, before any web content is downloaded. If an attempt is made to access a phishing site, the employee will be directed to a block page before any harm is done. DNS filters can also block malware downloads from sites that are not yet known to be malicious.
Employees are the weak link that are targeted by cybercriminals so it is important they are trained how to recognize phishing emails. You should provide security awareness training regularly to develop security aware culture in your organization. Over time, employees can be conditioned to respond correctly and report phishing threats to the security team. Also conduct phishing simulation exercises to make sure training has been effective. A failed phishing simulation allows you to identify a weak link and provide further training.
If all of the above defenses have failed, there is another layer that can keep your business protected: Multi-factor authentication. MFA requires another factor to be used before access to an email account or other system is provided. If an employee’s login credentials are disclosed in a phishing attack, MFA should stop those credentials from being used by a cybercriminal to access to gain access email accounts and other systems.
All of these layers are necessary to block today’s sophisticated phishing threats. It may seem like a lot of expense, but the above anti-phishing measures need not be expensive. TitanHQ can’t train your employees to be security titans, but through SpamTitan Email Security and WebTitan DNS filtering, phishing threats can be blocked.
by titanadmin | Dec 17, 2019 | Internet Security, Phishing & Email Spam, Spam Advice, Spam Software, Website Filtering |
IT professionals have long known that employees are a weak link in the security chain. Recent studies have confirmed this to be the case. Employees are poor at identifying phishing emails and other email-based threats and, to be fair on employees, many have received no training and phishing scams are becoming much more targeted and sophisticated.
The number of successful phishing attacks on businesses is difficult to determine, as many attacks go unreported, even when they result in the exposure of consumer data. In regulated industries, such as the healthcare industry in the United States, the picture is much clearer.
The Health Insurance Portability and Accountability Act – or HIPAA as it is better known – requires healthcare organizations to report breaches of patient information. Summaries of data breaches of 500 or more records are also made public and can be seen on the Department of Health and Human Services’ Office for Civil Rights data breach portal.
In 2019 alone, there have been at least 147 incidents of hacking of email accounts. The cost of those breaches is staggering. In those 147 incidents, the hacked email accounts contained the records of 2,762,691 individuals. According to the Ponemon Institute/IBM Security 2019 Cost of a Healthcare Data Breach report, the cost per exposed healthcare record is $423. Those breaches are therefore likely to have cost $1,168,618,293.
A recent study conducted by GetApp confirmed how often employees are fooled by phishing attacks in other industries. For the study, 714 individuals were surveyed from a range of businesses in the United States. Almost a quarter of those businesses have experienced at least one successful phishing attack and 43% of employees said that someone in their organization had clicked on a phishing email.
The aim of the study was to explore whether businesses were providing security awareness training to their employees to help them identify phishing emails. Only 27% of organizations did. It is therefore no surprise that employees often fall for phishing scams.
The provision of security awareness training, with a particular focus on phishing and social engineering, is vital. Even with layered defenses, some phishing emails will arrive in inboxes, so employees need to be taught the skills they need to help them identify email threats. Employees should then be tested by conducting phishing email simulations. That allows businesses to find out if the training has been taken on board. Without training and testing, employees will remain a liability. Over time their phishing identification shills will improve.
It is worth noting that security awareness training for employees is a requirement of HIPAA, yet many employees are still fooled. Training and phishing simulations can help reduce an organization’s susceptibility to phishing attacks, but employees, being human, will still make mistakes.
The solution is layered defenses. No one cybersecurity solution will block all phishing attempts, and certainly not without also blocking many legitimate email communications. Multiple solutions are therefore required.
It is essential for advanced email security defenses to be implemented to block phishing emails and make sure phishing and malspam (spam emails containing malware) never reach inboxes. That means an advanced spam filtering solution is a must.
SpamTitan for has been independently tested and shown to block in excess of 99.9% of spam emails and 100% of emails containing known malware. SpamTitan also blocks zero-day threats using a combination of advanced detection techniques. This is achieved through heuristic analyses, blacklists, trust scores, greylisting, sandboxing, DMARC, and SPF to name just a few.
SpamTitan has also been developed to compliment Office 365 security and provide a greater level of protection against phishing and other malicious email threats. It should be noted that Microsoft’s Exchange Online Protection was recently shown to allow 25% of phishing emails through.
Should phishing emails arrive in inboxes and be opened by end users, other controls are required to prevent clicks from resulting in malware infections or the theft of credentials. Here a web filtering solution such as WebTitan is important. When a link in an email is clicked, before the webpage is displayed, the URL and the content of the webpage is checked and the user is prevented from visiting the webpage if it, or its domain, is associated with phishing or malware distribution. Malware downloads can also be blocked from websites, even those with a high trust score. Together these solutions form the backbone of your phishing defenses. Further, these two solutions are quick and easy to implement, simple to use and maintain, and they are inexpensive.
Add antivirus protection, multi-factor authentication, and end user training, and you will be well protected from phishing and email and web-based malware attacks.
For further information on improving your defenses against phishing, spear phishing, and malware, give the TitanHQ team a call today.
If you are a managed service provider, contact the TitanHQ channel team and discover why TitanHQ is the leading provider of cloud-based email and web security solutions for MSPs serving the SMB market.
by titanadmin | Dec 16, 2019 | Industry News, Spam News, Spam Software |
Over the past 2 decades TitanHQ has been developing powerful cybersecurity solutions for SMBs and managed service providers (MSPs) that serve the SMB market. Naturally at TitanHQ we have great belief in our email security solution, SpamTitan. We believe it is the ideal spam filtering solution for SMBs and MSPs for preventing a myriad of email threats from reaching inboxes.
TitanHQ is the leading provider of cloud-based email security to MSPs serving the SMB market. We regularly receive positive feedback from MSPs and SMBs about how the solution has saved them hours of work compared to other email security solutions and has helped them improve email security and block more spam and stop malware and ransomware from reaching inboxes.
Positive feedback from end users proves we are getting it right and it inspires us to continue improving the solution to ensure it will keep on protecting our customers from malware, ransomware, viruses, botnets, and social engineering and phishing attacks for many years to come.
The positive feedback is not only provided to our engineers and customer service and sales teams. IT decision makers have posted highly positive reviews on the top business software review platforms and are letting other IT professionals know about their experiences implementing the solution, integrating it with their other cybersecurity solutions and management platforms, and what it is like to use SpamTitan on a daily basis.
In fact, across the different business review sites, SpamTitan has consistently received high scores. There is no other email security product on the market that has achieved such a wealth of positive reviews and feedback from end users.
Some of the positive reviews across the leading business software review sites are detailed below:
Gartner Peer Insights
Gartner Peer insights is one of the most highly respected review platforms from the world’s leading business advisory and research company. While Gartner strictly polices the review site, Gartner is unbiassed and has no hidden agenda. The review platform gives IT professionals the opportunity to give their honest feedback on software solutions that they have implemented to help other IT professionals save time and money in their search.
36 qualified users of SpamTitan have left reviews on the site and the solution has achieved highly positive feedback with an average user score of 4.7 out of 5.
“SpamTitan has been a very responsive vendor to work with, both during the sales process and with post-sales support. Tickets are responded to within several hours and often resolved within a day. The product itself is very MSP-friendly supporting delegation to client admins, multiple delivery pools, and attractive pricing. The catch rate is better than Exchange Online.”
Microsoft Team Lead in the Services Industry
“SpamTitan takes a little technical knowhow, but it’s powerful, flexible and affordable.” Director of IT and Telecom in the Healthcare Industry.
“SpamTitan is superb giving control back to the user and giving time back to IT staff. The product is amazing, it stopped 99% of spam and gives total control back to the user, it is web based and was easy to migrate to. The support and migration management from TitanHQ was brilliant.” IT Security Manager in the Manufacturing Industry.
G2 Crowd
G2 Crowd is one of the leading business software review sites. 139 verified users of SpamTitan have left reviews on the site and the solution has achieved an overall score of 4.6 out of 5. SpamTitan has been rated consistently highly in all rating categories, achieving 9.3 out of 10 for meets requirements and ease of doing business with, 9.2 for ease of setup and quality of support, 9.1 for ease of use, and 9.0 for ease of admin.
Additionally, each quarter, G2 Crowd compiles its Email Security Grid and rates solutions based on customer feedback and market presence. For four consecutive quarters, SpamTitan has been the Top Email Security Solution.
“I really like the customization that is available for this product. We have total control over the spam filter environment for all our customers. The environment is stable which is very important to us and our customers. The support staff was great when we were getting our environment configured. They were quick to reply to emails and reach out to assist us as needed. The spam filtering is top-notch and much better than other products we have used.”
Jeff Banks, Director Of Technology.
Antispam that is affordable, flexible and powerful.” Mike S, Director of IT and Telecommunications.
“Cloud Version is Great for Managed Service Providers.” Andrew B, Vice President.
“Minimizes our exposure to harmful malware and junk emails.” David C, Outreach Specialist.
Google Reviews
112 users of SpamTitan have taken the time to submit their feedback to Google Reviews. The solution is consistently given top marks by users and has achieved an overall review score of 4.9 out of 5.
Some of the positive feedback from users includes:
“TitanHQ is an excellent solution which ticks many boxes. It’s simple to setup, and gives a huge range of functionality all from within one place. My experience of the Support help desk has been great with a team that really do know their product. I highly recommend TitanHQ.” Chris Bell.
“The Titan Span filter is by far one of the best email filters I have ever used. It was simple to setup, it allows users to release their own emails from quarantine quick and easy.” Joseph Walsh.
“Great product. Spam reduced to almost zero and no user complaints. Configuration is simple and support is awesome. Love it!” George Homme.
Capterra
Capterra is a leading software review site that has been active for 20 years. The site has now been purchased by Gartner which moderates reviews on the site. Capterra includes more than 700 categories of software products and is one of the most highly respected business software review sites. It is relied upon by IT decision makers the world over.
SpamTitan has been reviewed by 379 users and has achieved an overall review score of 4.6 out of 5.
“It’s as close to “set it and forget it” as you can come in the IT field. Right out of the box support helped me set everything up in less than 20 minutes, no hardware to worry about, nothing like that. Literally all I have to do is check to see if something was blocked incorrectly once in a while, white list it, and done. I’ve been using spam titan for almost a year and in that time we have blocked over 200k spam/malicious emails for a 30 person company before they even hit employee mailboxes. I shut off the service for 48 hours just to make sure it easy legit, it was, and I haven’t shut it off again since. Whitelisting and blacklisting domains and specific emails are super easy. Support Staff are awesome and go into detail when resolving problems if they were to arise or even if you just have a question. They have always been friendly and courteous and super personable and have been some of the best people to work with in all my years doing IT.” Benjamin Jones, Director Of Information Technology.
“SpamTitan has saved me, saved my company time, and has some of the best support people around. It’s as close to “set it and forget it” as you can come in the IT field. Right out of the box support helped me set everything up in less than 20 minutes, no hardware to worry about, nothing like that. Literally all I have to do is check to see if something was blocked incorrectly once in a while, white list it, and done. I’ve been using spam titan for almost a year and in that time we have blocked over 200k spam/malicious emails for a 30 person company before they even hit employee mailboxes.” Benjamin J, Director of Information Technology.
Spiceworks
Members of the Spiceworks community have also rated SpamTitan highly. The solution has been reviewed by 56 users and has an overall rating of 4.6 out of 5.
Software Advice
The software review site Software Advice includes 350 reviews of SpamTitan from business users and has achieved an average score of 5.58 out of 5.
SpamFilterReviews
According to SpamFilterReviews, SpamTitan is the top-rated spam filtering solution on the site with a score of 4.9 out of 5.
by titanadmin | Dec 13, 2019 | Industry News, Internet Security, Network Security, Spam News |
Cyberattacks on managed service providers have been increasing over the past few months and they are now a key target for hackers. If a hacker can gain access to the systems of a managed service provider, their remote administration tools can be used to launch attacks on their clients.
There have been several major cyberattacks on managed services providers in the past few weeks, with nation-state-backed hacking groups targeting MSPs serving enterprises and ransomware gangs are conducting attacks on MSPs serving small and medium-sized businesses.
Three major cyberattacks on managed service providers serving healthcare organizations in the United States have been reported in the past two months. All three have affected more than 100 healthcare clients and one impacted 400.
In late November, the Milwaukee-based managed IT service provider, Virtual Care Provider Inc., was attacked with Ryuk ransomware. The attack started on November 17, 2019, and affected all of its clients’ data. Around 110 nursing homes and acute care facilities were prevented from accessing their patients’ medical records. The consequences for its clients were dire. Assisted living facilities and nursing homes were prevented from billing for Medicaid, which meant essential funding was not provided and nursing homes were prevented from ordering essential drugs for patients. Virtual Care Provider was issued with a $14 million ransom demand, which the company could not afford to pay. The managed service provider had around 20% of its services affected and had to rebuild around 100 servers.
The ransomware was deployed as a secondary payload by the TrickBot Trojan. TrickBot had been installed on its network 14 months previously via a malicious email attachment.
A few weeks later, a Colorado-based managed service provider serving dental practices was attacked with ransomware. Complete Technology Solutions was infected with a ransomware variant called Sodinokibi. First, the MSP was attacked, and then its remote administration tools were used to deploy ransomware on the networks of more than 100 dental practices. A ransom demand of $700,000 was issued, which the MSP refused to pay. Its clients are now having to pay the attackers for the keys to decrypt their files. Only a few that had backups stored off the network were able to recover without paying the ransom.
This is the second such attack to affect a company serving the dental industry. The dental record backup service provider, PerCSoft, was also attacked with Sodinokibi ransomware. That attack affected approximately 400 dental practices. CyrusOne was also attacked with Sodinokibi ransomware and its managed services division and six of its clients were affected.
It is not only ransomware that is being used in the attacks. Nation-state threat groups such as APT10 are also targeting MSPs. Their aims are different. The attacks are being conducted to gain access to the intellectual property of their enterprise customers.
As cyberattacks on managed service providers increase, MSPs must ensure that they have adequate defenses in place to keep the hackers at bay. This is an area where TitanHQ can help. TitanHQ is the leading provider of cloud-based email and web security solutions for managed service providers that serve the SMB market.
TitanHQ offers a trio of solutions for MSPs under the TitanShield program. SpamTitan email security is a powerful cloud-based solution that keeps inboxes free of spam, phishing emails, and malware. SpamTitan incorporates SPF and DMARC to block email impersonation attacks, dual antivirus engines to detect known malware threats, and heuristics and sandboxing to identify and block zero-day threats.
WebTitan Cloud is a 100% cloud-based DNS filtering solution that works seamlessly with SpamTitan to block web-based phishing attacks and malware downloads. The solution allows you to monitor and identify malicious threats in real-time and includes AI-driven protection against active and emerging phishing URLs, including zero-minute threats.
The third solution is ArcTitan, a cloud-based email archiving solution that provides protection against data loss and helps MSPs and their clients meet their compliance obligations. ArcTitan serves as a black box flight recorder for email and stores email data securely in the cloud on Replicated Persistent Storage on AWS S3. When emails need to be searched and recovered, the searches are lightning-fast. ArcTitan can search up to 30 million emails a second.
ArcTitan has recently been moved to a brand new system, with the service delivered as a highly available, self-healing horizontally scaled Kubernetes cluster. Within that cluster are many different components working in harmony together, but independently. Should any component go down, that component can be taken offline and repaired with no impact on the others, ensuring a much more reliable service with minimal or no disruption during an outage. With ArcTitan, email is protected from cyberattacks.
These solutions are not only an ideal for improving the security posture of MSP clients, they can help to ensure that MSP systems are protected from attack. All TitanHQ solutions are quick and easy to implement, have a low management overhead, and are API-driven so they can easily be incorporated into MSP’s remote management and monitoring systems.
To find out more about the TitanShield program for managed service providers and to discover how TitanHQ’s cybersecurity solutions can improve yours and your clients’ security posture, give the TitanHQ channel team a call today.
by titanadmin | Nov 30, 2019 | Email Scams, Internet Security, Spam Advice, Spam News |
Recent research has highlighted just how important it is for businesses to implement a range of defenses to ensure phishing emails are not delivered to inboxes and how business phishing protections are failing.
The studies were conducted to determine how likely employees are to click on phishing emails that arrive in their inboxes. Alarmingly, one study indicated almost three quarters of employees were fooled by a phishing test and provided their credentials to the attacker. In this case, the attacker was the consultancy firm Coalfire.
71% of the 525 businesses that were tested had at least one employee disclose login credentials in the phishing test, compared to 63% last year. At 20% of businesses, more than half of the employees who were tested fell for the phishing scam, compared to 10% last year.
A second study conducted by GetApp revealed a quarter of 714 surveyed businesses said they had at least one employee who responded to a phishing attack and disclosed their login credentials and 43% of businesses had employees that had clicked on phishing emails. The study also revealed only 27% of businesses provide security awareness training to employees, only 30% conduct phishing simulations, and 36% do not have multi-factor authentication in place on email.
The Importance of Layered Phishing Defenses
To mount an effective defense against phishing and other cyberattacks, a defense in depth approach to security is required.
With layered defenses, businesses are not replying on a single solution to block phishing attacks. Multiple defenses are put in place with the layers overlapping. If one measure proves to be ineffective at blocking a phishing email, others are in place to provide protection.
One area where many businesses fail is relying on Office 365 anti-phishing controls. A study by Avanan showed Office 365 phishing defenses to be effective at blocking most spam emails, but 25% of phishing emails were delivered to inboxes.
What is required is an advanced anti-spam and anti-phishing platform that can be layered on top of Office 365 to ensure that these phishing emails are blocked. SpamTitan can be seamlessly implemented in Office 365 environments and provides superior protection against phishing and malware attacks. SpamTitan blocks more than 99.9% of spam and phishing emails, 100% of known malware, and incorporates a host of features to identify zero-day threats.
As good as SpamTitan is at blocking email threats, other layers should be implemented to block phishing attacks. If a phishing email arrives in an inbox, a web filter will provide protection by blocking attempts by employees to visit phishing websites and sites hosting malware. WebTitan is a powerful DNS filtering solution that protects against the web-based element of phishing attacks. WebTitan adds an extra layer to phishing defenses and will block attempts by employees to visit malicious sites.
If an attacker succeeds in obtaining the credentials of an employee, it is important that those credentials cannot be used to gain access to the account. That protection is provided by multi-factor authentication. Multi-factor authentication is not infallible, but it will prevent stolen credentials from being used to access accounts in the majority of cases.
Security awareness training is also vital. Employees are the last line of defense and that defensive line will be tested. If employees are not trained how to identify phishing emails and other email security threats, they cannot be expected to recognize threats when they land in inboxes. An annual training session is no longer enough, considering how many phishing attacks are conducted on businesses and how sophisticated the attacks are becoming.
Security awareness training should consist of an annual training session with regular refresher training sessions throughout the year. Employees should be kept up to date on the latest tactics being used by cybercriminals to help them identify new scam emails that may bypass email security defenses. Phishing simulation exercises are also important. If these simulations are not conducted, businesses will have no idea how effective their training sessions have been, and which employees have not taken the training on board.
by titanadmin | Nov 27, 2019 | Phishing & Email Spam, Spam News |
A new phishing campaign has been detected that is targeting Office 365 admins, whose accounts are far more valuable to cybercriminals than standard Office 365 accounts.
A standard Office 365 email account can used for spamming or conducting further phishing attacks on the organization or business contacts. However, there is a problem. When the account is used for phishing, the sent messages are likely to be noticed by the user. Failed delivery messages will also arrive in the user’s inbox. The account may only be able to be used for a short time before an account compromise is detected.
The attackers targeting Office 365 admins aim to compromise the entire domain. Office 365 admins can create new accounts on the domain, which are then used for phishing. Since the only person using that account is the attacker, it is likely the malicious actions will not be noticed, at least not as quickly. The only person who will see the failed delivery messages and sent emails is the attacker.
The newly created account abuses trust in the business domain. Any individual to receive such a phishing message may mistakenly believe the email is a legitimate message from the company. The messages also take advantage of the reputation of a business. Since the business domain will have been used only to send legitimate messages, the domain will have a high trust score. That makes it far more likely that the emails being sent from the new account will be delivered to inboxes and will not be picked up by Office 365 spam filters. The Office 365 admin may also have access to all email accounts on the domain, which will allow the attacker to steal a huge amount of email data.
In theory, Office 365 admins should be better at identifying phishing emails than other employees in the organization as they usually work in the IT department; however, these emails are very realistic and will likely fool many Office 365 admins.
The lure being used is credible. The emails appear to have been sent by Microsoft and include the Microsoft and Office 365 logos. The emails claim that the organization’s Office 365 Business Essentials invoice is ready. The user is told to sign into the Office 365 admin center to update their payment information, set their Message Center preferences, and edit their release preferences or join First Release and set these up if they have not done so already. The emails include an unsubscribe option and are signed by Microsoft and include the correct contact information. The emails also link to Microsoft’s privacy statement.
The embedded hyperlinks in the emails link to an attacker-controlled domain that is a carbon copy of the official Microsoft login page. If the user’s credentials are entered, they are captured by the attacker.
This campaign highlights how important it is to have layered email security defenses in place to block phishing attacks. Many phishing emails bypass standard Office 365 anti-phishing controls so additional protection is required.
An advanced anti-phishing solution such as SpamTitan should be layered on top of Office 365 to provide greater protection against sophisticated phishing attacks. Approximately 25% of all phishing emails bypass standard Office 365 phishing protections.
Another anti-phishing layer that many businesses have yet to implement is a web filter. A web filter, such as WebTitan, provides protection when messages are delivered to inboxes, as it blocks attempts by employees to visit phishing websites. When a link to a known phishing website is clicked, or the user attempts to visit a questionable domain, they will be directed to a block page and the phishing attack will be blocked.
by titanadmin | Nov 26, 2019 | Email Scams, Phishing & Email Spam, Spam Advice, Spam News, Spam Software |
The aim of this post is to provide you with some easy to adopt email security best practices that will greatly improve your organization’s security posture.
Email is the Most Common Attack Vector!
It is a certainty that business email systems will be attacked so email security measures must be implemented. The best form of email security is to do away with email altogether, but since businesses rely on email to communicate with customers, partners, and suppliers, that simply isn’t an option.
Email not only makes it easy to communicate with the people you need to for your business to operate, it also allows cybercriminals to easily communicate with your employees and conduct phishing attacks, spread malware and, if a corporate email account is compromised, communicate with your customers, partners and suppliers.
Email security is therefore essential, but there is no single solution that will protect the email channel. A spam filtering solution will stop the majority of spam and malicious email from reaching inboxes, but it will not block 100% of unwanted emails, no matter what solution you implement. The key to robust email security is layered defenses. If one defensive measure fails, others are in place that will provide protection.
You need a combination of technical, physical, and administrative safeguards to secure your email. Unfortunately, there is no one-size-fits-all approach that can be adopted to secure the email channel but there are email security best practices that you can adopt that will improve your security posture and make it much harder for cybercriminals to succeed.
With this in mind, we have outlined some of the most important email security best practices for your business and your employees to adopt.
Email Security Best Practices to Implement Immediately
Cybercriminals will attempt to send malware and ransomware via email, and phishing tactics will be used to steal sensitive information such as login credentials, so it is important to be prepared. Listed below are 8 email security best practices that will help you keep your email system secure. If you have not yet implemented any of these best practices, or have only done so partially, now is the time to make some changes.
Develop a Cybersecurity Plan for Your Business
We have included this as the first best practice because it is so important. It is essential for you to develop a comprehensive cybersecurity plan for your entire organization as not all threats arrive via email. Attacks come from all angles and improving email security is only one of the steps you need to take to improve your overall cybersecurity posture.
There are many resources available to help you develop a cybersecurity plan that addresses all cyber risks. The Federal Communications Commission has developed a Cyberplanner to help with the creation of a custom cybersecurity plan and the Department of Homeland Security’s Cybersecurity and Infrastructure Security Agency (CISA) has recently issued a Cyber Essentials Guide for Small Businesses and Governments. Take advantage of these and other resources to develop an effective cybersecurity plan.
Implement an Advanced Spam Filtering Solution
A spam filter serves as a semi-permeable membrane that prevents email threats from being delivered to inboxes and lets genuine emails pass through unimpeded. This is the single most important security measure to implement to protect against email threats and productivity-draining spam.
If you use Office 365 you will already have some protection, as Office 365 includes a spam filter and anti-virus software, but it falls short on phishing protection and will not block zero-day malware threats. You need layered defenses to secure email which means a third-party spam filter should be used on top of Office 365. Research from Avanan showed that 25% of phishing emails bypass Office 365 defenses.
There are many spam filtering services for SMBs, but for all-round protection against known and zero-day threats, ease of implementation, ease of use, and price, SpamTitan is the best choice for SMBs.
Ensure Your Anti-Virus Solution Scans Incoming Emails
You will no doubt have anti-virus software in place, but does it scan incoming emails? Email is one of the main ways that malware is delivered, so anti-virus software for email is a must. This does not necessarily mean you need a different antivirus solution. Your existing solution may have that functionality. Your spam filter is also likely to include AV protection. For example, SpamTitan incorporates dual anti-virus engines for greater protection and a sandbox where email attachments are analyzed for malicious actions. The email sandbox is used to detect and block zero-day malware – New, never-before-seen malware variants that have yet to have their signatures incorporated into AV engines.
Create and Enforce Password Policies
Another obvious email security best practice is to create a password policy that requires strong passwords to be set. There is no point in creating a password policy if it is not enforced. Make sure you implement a control measure to prevent weak passwords from being set. Weak passwords (password, 123456, or dictionary words for example) are easy to remember but also easy to guess. Consider that cybercriminals are not sitting at a computer guessing passwords one at a time. Automation tools are used that make thousands of password guesses a minute. It doesn’t take long to guess a weak password! You should also make sure rate limiting is applied to block an IP from logging in after a set number of failed login attempts.
It is a good best practice to require a password of at least 8 characters to be set, with a combination of upper- and lower-case letters, numbers, and symbols, and to block the use of dictionary words. Consider allowing long passphrases to be used as these are easier for employees to remember. Check the National Institute of Science and Technology (NIST) advice on secure password practices if you are unsure about creating a password policy.
Implement DMARC to Stop Email Impersonation Attacks and Domain Abuse
DMARC, or Domain-based Message Authentication, Reporting & Conformance to give it its full name, is an email protocol that uses Sender Policy Framework (SPF) and DomainKeys Identified Mail (DKIM) to determine whether an email is authentic.
By creating a DMARC record you are preventing unauthorized individuals from sending messages from your domain. DMARC also lets you know who is sending messages from your domain, and it lets you set a policy to determine what happens to messages that are not authenticated, I.e. quarantine them or reject them. Some email security solutions, such as SpamTitan, incorporate DMARC authentication.
Not only DMARC help you block email impersonation attacks, it also prevents abuse of your domain. Your DMARC record tells receiving email servers not to accept messages sent from authenticated users, thus helping protect your brand.
Implement Multi-Factor Authentication
Multi-factor authentication is yet another layer you can add to your anti-phishing defenses. Multi-factor authentication, as the name suggests, means more than one method is used to authenticate a user. The first factor is usually a password. A second factor is also required, which is something a person knows or possesses. This could be a mobile phone, to which a one-time PIN code is sent, or a token on a trusted device.
This safeguard is vital. If a password is obtained, in a phishing attack for example, the password alone will not grant access to the email account without an additional factor being provided. A combination of a password, token, and one-time PIN is a good combination.
Train Your Employees and Train Them Again
No matter how tech savvy your employees appear to be, assume they known nothing about cybersecurity. They will certainly not routinely stick to email security best practices unless you train them to do so and then hammer the message home.
Before letting any employee have access to email, you should provide security awareness training. Your training should cover email security best practices such as never opening email attachments from unknown senders, never enabling content in documents unless the document has been verified as legitimate, and never to click hyperlinks in emails or send highly sensitive information such as passwords via email.
You must also train your employees how to recognize phishing emails and other malicious messages and tell them what to do when suspicious emails are received. Anyone with access to email or a computer must be provided with security awareness training, from the CEO down.
One training session is not enough. Even an annual training session is no longer sufficient. You should be providing regular training, be sending cybersecurity newsletters warning about the latest threats, and using other tools to help create a security culture in your organization.
Conduct Phishing Awareness Simulation Exercises
You have provided training, but how do you know if it has been effective? The only way to tell is to conduct tests and that is easiest with phishing simulation exercises. These are dummy phishing emails that are sent to employees when they are not expecting them to see how they respond. You maybe surprised at how many employees respond and disclose sensitive information, open attachments, or click links in the emails.
The aim of these emails is to identify people that have not taken their training on board. The idea is not to punish those employees, but to tell you who needs further training. There are several companies that can assist you with these exercises. Some even offer free phishing simulation emails for SMBs.
TitanHQ is Here to Help!
TitanHQ has developed SpamTitan to be easy for SMBs to implement, use, and maintain. It requires no hardware, no software, and all filtering takes place in the cloud. Not only does SpamTitan offer excellent protection against the full range of email-based threats, it is also one of the lowest cost solutions for SMBs to implement.
Give the TitanHQ team a call today for more information on SpamTitan and to find out about how you can also protect your business from web-based threats and meet your compliance requirements for email.
by titanadmin | Nov 26, 2019 | Industry News, Phishing & Email Spam, Spam Advice, Spam Software |
SMBs and Managed Service Providers (MSPs) that serve the SMB market have many spam filtering services to choose from. In this post perform a VadeSecure vs SpamTitan Email Security comparison to help you decide on the best solution to meet the needs of your business.
Who are VadeSecure?
VadeSecure is a French company that was founded in 2009. The company has developed a predictive email defense solution to protect businesses from email-based threats and spam email, and also consumers through their ISPs. The company has yet to make great inroads in the MSP market, although that is part of the company’s plan, having recently raised $79 million in venture capital to help them achieve this aim.
SpamTitan Email Security from TitanHQ
TitanHQ is the leading provider of cloud-based email and web security solutions for MSPs that serve the SMB market. TitanHQ has more than 2 decades of experience in email and web security and has developed two award winning solutions for MSPs – WebTitan (Web Security) and SpamTitan Email Security. Here we will focus on SpamTitan Email Security.
VadeSecure vs SpamTitan Email Security
Take a quick look at VadeSecure and SpamTitan Email Security and you may think that both solutions are very similar, and in some respects they are. Both are cloud-based email security solutions that have been designed to block email threats and keep inboxes free from spam and malicious messages and attachments. Both solutions have been developed to provide an additional security layer for Office365 to block the many spam and malicious messages that bypass O365 security controls.
However, there are some very important differences between the solutions as far as MSPs are concerned. VadeSecure has been developed solely for the Telco market, but MSPs have unique requirements that are not well catered to. A deeper dive into the products and a more thorough comparison of VadeSecure vs SpamTitan Email Security from an MSP perspective reveals the two solutions are very different products.
SpamTitan is very much MSP focused. Over time, with the increased investment, VadeSecure may become a more MSP friendly solution, but as it stands VadeSecure and SpamTitan Email Security are not equivalent solutions.
Comparison of VadeSecure and SpamTitan Email Security for MSPs
SpamTitan Email Security has been developed by MSPs for MSPs. SpamTitan Email Security is therefore a very MSP-focused product, which incorporates many MSP-friendly features. SpamTitan is a true multi-tenant solution. With SpamTitan Email Security, MSPs are given a multi-tenancy view of all customers with multiple management roles. This allows MSPs to easily monitor all customer deployments and the trial-base, assess the health of those deployments, view activity volumes across your entire customer base, and quickly identify any issues that need to be addressed. VadeSecure lacks this customer-wide view of the system and does not integrate with RMMs or PSAs.
Configurability and Customization Potential
Configurability is also a key consideration. VadeSecure is not easily configurable to meet your needs. For instance, it does not support custom rules, so you have to use Office 365 Exchange admin functionality for configuration. In a similar vein, the potential for customization is limited with VadeSecure. With SpamTitan Email Security, there is plenty of scope for customization. You can create custom rules to meet the needs of your customer base thanks to highly granular controls that can be applied to domains, groups, or individual users. This level of granularity is important, as it allows you to carefully configure the solution to meet the needs of each client. You can tailor the solution to suit the risk tolerance of each individual client and adopt a more aggressive or more permissive approach on a per client basis and minimize false positives and false negatives. VadeSecure lacks the granularity to allow this for each customer.
Management and Reporting
You are implementing email security to provide your customers with greater security, but you need to make sure the solution remains effective over time. You will therefore need to identify issues as they arise and perform tweaks to continue to protect your clients to the highest degree. To achieve this, you need highly granular reports. Without them you will not have the visibility you need. SpamTitan’s suite of pre-configured and customizable reports give you full visibility into your deployments to allow you to quickly identify and correct any issues.
You can also generate reports (manually or automatically) that you can send to your clients to show them how effective the solution is, the threats that are being blocked, and why continued protection is essential. With VadeSecure you lack this visibility and cannot find out what has been blocked for end users or obtain detailed information on spam emails and threats. Client management is also more difficult with VadeSecure. MSPs need to login to each client’s Office 365 environment for management, which makes reporting much more time consuming.
Revenue Potential and Margins
Because SpamTitan allows MSPs to customize their deployments, MSPs have superior management capabilities and can offer clients greater value, which means greater margin potential for MSPs. It also makes it harder for clients to switch providers as their MSP is more of a strategic partner rather than just an IT service provider.
With TitanHQ there is also greater potential to make more margin by cross selling other services. MSPs that sign up with TitanHQ and join the TitanShield program have access to two other revenue generating solutions: WebTitan DNS filtering and ArcTitan Email Archiving. These allow you to maximize monthly recurring revenue with each client. Additional revenue-generating solutions are not available with VadeSecure.
VadeSecure Vs SpamTitan Email Security Pricing
Currently, pricing with VadeSecure is complex and the solution is expensive for MSPs. VadeSecure is charged on a per module basis, which means you need to factor in a lot of additional costs, such as anti-virus protection and GreyMail which are not included as standard. With SpamTitan there is one flat fee that includes all features of the solution. TitanHQ pricing is totally transparent and there are no hidden extras.
After speaking with customers that have tried VadeSecure, we have learned that the total number of users are not aggregated into the MSP discount with VadeSecure. You could have 100 x 10-seat licenses (1,000 users), but VadeSecure pays at 10 seats each and not the 1,000 seats overall. In contrast, TitanHQ’s appreciates how MSPs work and has developed a flexible pricing policy accordingly.
Quick Comparison of Features
In the image below we have compared the basic features of both SpamTitan and VadeSecure as a quick reference to show you some of the key differences between VadeSecure and SpamTitan Email Security.

MSPs that serve customers with Office 365 environments should adopt a layered approach to security and should not rely on the anti-spam and anti-phishing defenses incorporated into Office 365. Additional layers are required to better protect clients, which will mean you spend less time on support and remediating phishing attacks.
TitanHQ can provide two additional layers to your security stack: SpamTitan and WebTitan, both of which work seamlessly together to protect against all email and web-based threats.
To find out more about these solutions, how you can reduce the cost of email security and web security for your customers while earning a profitable margin, contact the TitanHQ team today and ask to speak to the channel team.
by titanadmin | Nov 18, 2019 | Internet Security, Network Security, Spam Advice, Spam Software, Website Filtering |
Cybercriminals are inventive and their attacks are becoming increasingly sophisticated. To help ensure you are prepared and can defend your business against these attacks, we have listed the top 10 cybersecurity threats your business is likely to face, along with some tips to help you prevent a costly data breach.
Cybercriminals are not just trying to attack large enterprises. Sure, a cyberattack on a large healthcare system or blue-chip company can be incredibly rewarding, but the defenses they have in place make attacks very difficult. SMBs on the other hand have far fewer resources to devote to cybersecurity and as a result they are easier to attack. The potential rewards may not be as great, but attacks are more likely to succeed which means a better return on effort. That is why so many SMBs are now being attacked.
There is a myriad of ways that a company can be attacked, and the tactics, techniques and procedures used by cybercriminals are constantly changing. The top 10 cybersecurity threats listed below include the main attack vectors that need to be blocked and will serve as a good starting point on which you can build a robust cybersecurity program.
Top 10 Cybersecurity Threats Faced by SMBs
We have listed the top 10 cybersecurity threats that SMBs need to defend against. All the threats listed below need to be addressed as any one of them could easily result in a costly data breach, data loss, or could cripple your business. Some of the threats listed below will be harder to address than others, and it will take time for your cybersecurity defenses to mature. The important thing is to start the ball rolling and address as many of these areas as soon as possible.
Human Error and Insider Threats
We have listed human error first, as it doesn’t matter what hardware and software solutions you implement, human error can easily undo much of your good work. Mistakes will be made by employees on occasion. What you need to do is reduce the potential for errors and limit the harm that can be caused.
Developing robust policies and procedures and providing training will help to ensure that your employees know how to act and more importantly, how not to.
Mistakes are not the only thing you need to take steps to try to prevent. There may also be individuals on your payroll who will take advantage of poor security for personal gain. You will also need to tackle the problem of insider threats and make it harder for rogue employees to cause harm and steal data. The measures listed below will help address threats from within and reduce risk.
- Passwords
- Enforce the use of strong passwords but make it easier for your employees to remember them so they don’t try to circumvent your password policy or, heaven forbid, write their passwords down. Implement a password manager to store their passwords so they only have one password or pass phrase to remember.
- Rule of Least Privilege
- It is obvious, but often overlooked. Don’t give employees access to resources they do not need for their day-to-day work duties. If their credentials are compromised, this will limit the harm caused. It will also limit the harm that can be caused by rogue employees.
- Block the Use of USB Devices
- USB devices make it easy for rogue employees to steal data and for malware to be accidentally or deliberately be introduced. Implement technical controls to prevent USB devices from being connected, and if they are required for work purposes only give permission to certain individuals to use them. Ideally, use more secure methods of transferring or storing data.
- Monitor Employee Activity
- If rogue employees are stealing data, you are only likely to find out if you are monitoring their computer activity. Similarly, if credentials are compromised, system logs will highlight any suspicious activity. Make sure logs are created and monitored. Consider using a security information and event management (SIEM) solution to automate this as much as possible.
- Terminate Access at Point of Termination
- Terminating an employee? Terminate their access to your systems at the point of termination. It is surprising how often employee access rights are not terminated for days, weeks, or even months after an employee has left the company.
We will cover some more important safeguards to implement to protect against user error in the following 9 SMB cybersecurity threats.
Phishing and Social Engineering Attacks
Phishing is arguably the biggest cybersecurity threat faced by SMBs. Phishing is the use of social engineering techniques to persuade people to divulge sensitive information or take an action such as installing malware or ransomware. This is most commonly achieved via email, but can also occur via text messages, social media websites, or over the telephone.
Do not assume that your employees have common sense and know not to open email attachments from unknown individuals or respond to enticing offers from legal representatives of Nigerian princes. You must train your employees and teach cybersecurity best practices and show them how to identify phishing emails. Refresher training should be provided at regular intervals and you should conduct phishing simulation exercises (which can largely be automated) to find out who has taken the training on board and who is a liability that needs further training.
Employees are the last line of defense. You need a layer of security above your employees to make sure their security awareness training is never required. That means an advanced anti-spam solution needs to be in place to block threats before they reach inboxes. If you use Office 365, you should still implement an antispam solution. A recent study by Avanan revealed 25% of phishing emails bypass Office 365 antispam defenses.
Another layer of protection should also be implemented to protect against phishing: Multi-factor authentication. This is the use of an additional authentication factor that will kick into action if an attempt is made to use credentials from an untrusted device or location. If credentials are compromised in a phishing attack, multi-factor authentication should stop them from being used to gain access to email accounts, computers, or network resources.
Malware and Ransomware
Malware, viruses, ransomware, spyware, Trojans, worms, botnets, and cryptocurrency miners are all serious threats that you must take steps to block. It goes without saying, but we will say it none the less, you need to have antivirus software installed on all endpoints and your servers.
Malware can be installed in many ways. As previously mentioned, blocking USB devices is important and spam filtering software with sandboxing will protect you from email-based attacks. Most malware infections now occur via the internet, so a web filtering solution is also important. This will also add an extra layer to your phishing defenses. A web filter will block drive-by malware downloads, prevent employees from visiting malicious sites (including phishing websites) and also allows you to enforce your internet usage policies. A DNS filtering solution is the best choice. All filtering takes place in the cloud before any content is downloaded and it will not add to your patching burden.
Shadow IT
Shadow IT – The term given for any hardware or software in use that has not been authorized by your IT department. This could be a portable storage device such as a zip drive, a VPN client to bypass your web filter, an application to help with work tasks, or all manner of other software. It is surprising to find exactly how many of these programs are installed on users’ devices when IT support staff are called upon to sort out a problem!
So, what is the problem? Anything installed without authorization is a potential security and compliance risk. Your security team has no control over patching, and vulnerabilities in those applications could easily go addressed for months and give hackers an easy entry point into your network. Fake applications could be downloaded that are really malware, software packages often include a host of potentially unwanted programs and spyware, and any data stored in these applications could be transmitted to unsecure locations. Those applications and data contained therein are also unlikely to be backed up by the IT department. If anything happens, data can easily be lost.
Unpatched Software
The importance of prompt patching cannot be understated. Vulnerabilities exist in all software solutions. Sooner or later those vulnerabilities will be found, and exploits will be developed to take advantage. Security researchers are constantly looking for flaws that could potentially be exploited by threat actors to gain access to sensitive information, install malware, or remotely execute code. When these flaws are identified and patches are released, they need to be applied promptly. Oftentimes, vulnerabilities are being actively exploited by the time a patch is released. It is essential for these vulnerabilities to be addressed as soon as possible and for all software to be kept up to date.
When software or operating systems are approaching end of life, you must upgrade. When patches stop being issued and software is unsupported, any vulnerabilities will remain unaddressed and can easily be exploited.
Out of Date Hardware
Not all vulnerabilities come from out of date software. The hardware you use can also introduce risks. You must keep an inventory of all your hardware, so nothing slips through the cracks. Firmware updates should be applied as soon as it is made available and you should monitor for any devices that are approaching end of life. If your devices do not support the latest operating systems, then it is time to replace your hardware. This will naturally come at a cost, but so do cyberattacks and data breaches.
Unsecured IoT Devices
The Internet-of-Things offers convenience but IoT devices are a potential liability. IoT devices can send, store or transmit data so they must be be secured.
Unfortunately, in the hurry to connect everything to the internet device manufacturers often overlook security as do users of these devices. Take security cameras for instance. You may be able to access your cameras remotely, but you may not be the only person who can. If your security cameras are hacked, thieves could see what you have, where it is located, and where and when security is lax. There have been cases of security cameras being hacked due to the failure to change default credentials for remote management.
Ensure you change the default credentials on the devices and use strong passwords. Keep the devices up to date, and if the devices need to connect the network, make sure they are isolated from other resources. Cybercriminals can also take advantage of flaws in the applications to which these IoT devices connect. They must also be kept up to date.
Man-in-the-Middle Attacks and Public Wi-Fi
A man-in-the-middle (MITM) attack is an attack scenario where communications between two individuals (or one individual and a website or network) are intercepted and potentially altered. An employee may believe they are communicating securely, when everything they are saying or doing is being seen or recorded. An attacker could even control the conversation between two people and be communicating with each separately while both individuals believe they are communicating with each other. This method of attack most commonly occurs through unsecured Wi-Fi hotspots or evil twin hotspots – Fake Wi-Fi hotspots set up in coffee shops, airports, and any other location where free Wi-Fi is offered.
If you have remote workers, you need to take steps to ensure that all communications are kept private. This can be achieved in two main ways. By making sure employees use a secure VPN that encrypts their communications over public or unsecured Wi-Fi networks and also by implementing a DNS filtering solution. The DNS filtering solution provides the same protection for remote workers as it does for on-premises workers and will prevent malware downloads and employees from accessing malicious websites.
Mobile Security Threats
There is no denying the convenience of mobile devices (laptops, tablets, smartphones). They allow workers to be instantly contacted and lets them work from any location. Mobile devices improve employee mobility, can lead to greater employee satisfaction, and will help you to boost productivity. However, the devices also introduce new risks. Whether you supply these devices or operate a BYOD policy, you need to implement a range of security controls to ensure those risks are managed.
You need to make sure you know of every device that you allow to connect to the network. A mobile device security solution can help you gain visibility into mobile device use and allow you to control your applications and data.
You should ensure the devices have security controls applied, can only access your network via secure channels (VPN), ensure the devices are covered by a DNS filtering solution, and any work data stored on the devices needs to be encrypted.
Remote Desktop Protocol
Remote desktop protocol (RDP) allows employees remotely connect to your computers and servers when they are not in the office and lets your managed service provider quickly sort out your problems and maintain your systems without having to pay a visit. RDP also gives hackers an easy way to gain access your computers, servers, and steal data or install malware. Do you need RDP enabled? If not, disable it. Does it need to be used internally only? Make sure that RDP is not exposed to the internet.
If you do need RDP, then you need to exercise extreme caution. Make sure that users can only connect via a VPN or set firewall rules. Limit the individuals who have permissions to use RDP, ensure strong passwords are set, and that rate limiting is implemented to protect against brute force attacks. Also use multi-factor authentication.
Stolen RDP credentials are often used by hackers to gain access to systems, brute force attempts are often conducted, and vulnerabilities in RDP that have not been patched are frequently exploited. This is one of the main ways that ransomware is installed.
These are just the top 10 cybersecurity threats faced by SMBs. There are many more risks that need to be identified and mitigated to ensure you are protected. However, by addressing the above issues you will have already made it much harder for hackers and cybercriminals to do your business harm.
TitanHQ is Here to Help!
TitanHQ can assist by providing you with advanced cybersecurity solutions to protect against several of the above listed top 10 cybersecurity threats and will the two most commonly used attack vectors – email and the web-based attacks. These solutions – SpamTitan and WebTitan – are 100% cloud based, easy to implement and maintain, and will provide superior protection against malware, ransomware, viruses, botnets, and phishing attacks.
Further, these powerful solutions are affordable for SMBs. You are likely to be surprised to find out how little these enterprise-grade security solutions will cost. If you are a managed service provider that services the SMB market, you should also get in touch. SpamTitan and WebTitan have been developed by MSPs for MSPs. There is a host of reasons why TitanHQ is the leading provider of cloud-based email and web security solutions to MSPs that service the SMB market!
Contact our friendly (and non-pushy) sales team today to find out more, book a product demo, and register for a free trial.
by titanadmin | Nov 18, 2019 | Industry News, Spam Software |
TitanHQ has announced that a new version of its award-winning cloud-based anti-spam service and anti-spam software has been released. SpamTitan v7.06 incorporates a new RESTapi to allow clients and partners to seamlessly integrate SpamTitan into their own systems.
The new version was released on November 12, 2019 and has automatically been applied to the cloud-based offering. Users of SpamTitan software will have had the latest version downloaded, although they will need to login to their UI to apply the update.
As part of the regular patching cycle, SpamTitan patches have been released to address reporting engine issues and patches and ISO/OVA images are now available. These have been released for several packages including OpenSSL, OpenSSH, PHP, ClamAV and sudo. The patches must also be applied manually by administrators on their appliance(s).
TitanHQ has had a busy 2019. The company has experienced 30% growth in 2019 and has just had its busiest ever quarter for MSP growth. The growth has been driven by demand from MSPs for easy to use email security and web security solutions to protect their SMB clients from the growing number of cybersecurity threats.
TitanHQ now has more than 2,200 MSP partners using its platform and the strong Q3 growth has continued in Q4 helped by the new “Margin Maker for MSPs” Q4 initiative.
“Implementing the RESTapi and encouraging API adoption are vital steps in our partnership expansion plans,” explained TitanHQ CEO, Ronan Kavanagh. “We have enjoyed a record-breaking growth and the latest enhancements and new features that have been added to SpamTitan will help to ensure growth in 2020 will continue at record levels.”
by titanadmin | Nov 11, 2019 | Email Scams, Phishing & Email Spam, Spam Advice, Spam News |
Phishers are constantly changing tactics and coming up with new ways to fool people into handing over their credentials or installing malware. New campaigns are being launched on a daily basis, with tried and tested lures such as fake package delivery notices, fake invoices and purchase orders, and collaboration requests all very common.
In a departure from these common phishing lures, one threat group has opted for a rarely seen lure, but one that has potential to be very effective: Fake court subpoenas. The emails use fear and urgency and are designed to get users to panic and click quickly.
This campaign has been running for a few weeks and is targeting users in the United Kingdom, although this scam could easily be adapted and used in attacks on users in other countries.
Many phishing scams have the goal of stealing credentials to allow email accounts or Office 365 accounts to be accessed. In this case, the aim of the attack is to spread information stealing malware called Predator the Thief.
The phishing emails appear to have been sent by the Ministry of Justice in the UK. The sender field has Ministry of Justice as the display name and the emails have the Ministry of Justice crest, although the actual email address suggests the email has come from the Department of Justice (DOJ).
The emails warn the user that they have been subpoenaed. They are supplied with a case number along with a date when they have been ordered to attend court.
The emails include a hyperlink which the user must click to find out details of the charge and the documents they will need to bring with them to court. Urgency is added by warning the recipient they only have 14 days to respond to provide notice, and that the court case will proceed without them if they do not respond.
The URL in the email is seemingly benign, as it links to Google Docs – a trusted website. Clicking the link will see the user first directed to Google Docs, then redirected to OneDrive. When the user arrives on the OneDrive site, a document is downloaded. That document contains a malicious macro that launches a PowerShell command that downloads Predator the Thief malware.
Predator the Thief is an information stealer that can take screenshots and steals email and FTP credentials, along with cryptocurrency wallets and browser information. In contrast to many browser information stealers, this malware variant doesn’t just target the main browsers, but a host of less popular browsers. Once information has been stolen, the malware cleans up and exits, which makes it harder for the infection to be detected.
Phishing scams such as this highlight the need for layered security. Naturally, an advanced anti-spam solution such as SpamTitan should be implemented to block these threats and ensure and ensure messages are not delivered to end users’ inboxes. SpamTitan also includes DMARC email authentication to block mail impersonation attempts and a sandbox where email attachments are analyzed for malicious actions.
SpamTItan blocks in excess of 99.9% of all malicious emails, but it is not possible to block 100% of threats no matter what email security solution you use. This is where another layer is required. WebTitan is a DNS filtering solution that blocks threats such as this at the point where a DNS lookup is performed. This allows malicious websites to be blocked before any content is downloaded. WebTitan can also be configured to block downloads of certain file types.
With these two solutions in place, your business will be well protected against phishing emails and web-based malware downloads.
by titanadmin | Oct 23, 2019 | Email Archiving, Industry News, Internet Security, Network Security, Spam Software, Website Filtering |
Q3, 2019 has seen TitanHQ register record-breaking growth in the MSP market with its busiest ever quarter for MSP sales. TitanHQ now has more than 2,200 MSP partners and its cloud-based email security, web security, and email archiving platforms are now used by more than 8,200 businesses around the world.
Many great success stories start from humble beginnings, and TitanHQ is no exception. The company started life as Copperfasten Technologies in 1999 and sold anti-spam appliances to local businesses from its Galway, Ireland base. The company then developed its own cybersecurity solutions, starting with the anti-spam and anti-phishing solution, SpamTitan.
The product portfolio grew to include WebTitan web filtering, a powerful DNS-based web security solution to protect businesses from the full range of internet threats. That was followed by the launch of ArcTitan, a cloud-based email archiving solution for businesses that eases their email storage and compliance burden.
That trio of core TitanHQ products has proven to be a massive hit with managed service providers, although not by accident. Many companies have developed innovative solutions for SMBs but have only realized the importance of the MSP market later on. Additional features are then added to appeal to MSPs. TitanHQ took a different approach. Its solutions were developed by MSPs for MSPs and MSPs were considered at every stage of product development. The result is a suite of security solutions tailor-made for MSPs.
This approach, along with cutting-edge technology and industry-leading customer support, has seen the company go from strength to strength and become the gold standard in email and web security and the leading global provider of cloud-based security solutions for MSPs servicing the SMB market.
Phishing attacks on businesses are soaring, new malware variants are being released at record levels, and the current ransomware epidemic is threatening to derail businesses. Many SMBs lack the internal resources to block these threats and turn to MSPs to provide the security they need.
To cope with the increased demand, MSPs need solutions with 100% cloud-based architecture that seamlessly integrate into their existing centralized management systems and are easy to implement, use, and maintain. Ideally, those solutions need to be flexible, have a range of hosting options, be available in white-label form to take MSP branding, and also include generous margins. That is a big ask, and many solutions only tick a few of those boxes. However, TitanHQ’s suite of solutions include all those features and more.
TitanHQ also offers extensive sales enablement and marketing support, world-class customer service, and each MSP has a dedicated account manager, engineers, and a support team to help them maximize their sales opportunities and really grow their businesses.
As part of the celebration of the Q3, 2019 MSP growth, TitanHQ has launched a new initiative to ensure Q4 will be an even bigger success.
On October 22, TitanHQ announced a new disruptive price package for a SpamTitan Email Security and WebTitan DNS filtering bundle at an exclusive once-in-a-lifetime price. The initiative has been called Margin Maker for MSPs and is intended to ensure MSPs build profitability instantly in Q4, 2019.
The two solutions are provided in two private clouds, customized to meet MSPs email and web security needs, and secure the most common attack vectors – email and the web. The package includes advanced protection for email, including Office 365 environments, complimented by WebTitan DNS filtering to block web-based threats and implement content control for on-premises and remote workers. These solutions are naturally provided with extensive sales enablement and marketing support.
The aim is to make TitanHQ’s email and web security platforms even more appealing to MSPs and to encourage MSPs to offer both SpamTitan email security and WebTitan web filtering to their clients and maximize revenues.
One MSP that is already boosting its profits and achieving increased, reliable recurring monthly revenues is UK-based OpalIT. The MSP has bases in Newcastle and Edinburgh and a 6,000+ customer base. Prior to joining the TitanShield program, OpalIT was offering its clients firewall filtering and email filtering with Barracuda and Vade. The company has now switched to TitanHQ’s cybersecurity bundle and is pushing SpamTitan Email Security, WebTitan DNS filtering, and ArcTitan email archiving to its clients and is reaping the rewards.
“Opal IT moved to TitanHQ because of our MSP focused solutions, ease of deployments, extensive APIs functionality and the increased margin they’re now making. Our cybersecurity bundle solutions allow MSPs to provide their downstream customers with a layered defense approach” said Rocco Donnino, EVP Strategic Alliances, TitanHQ.
If you are a managed service provider, now is the perfect time to sign up with TitanHQ. Come and meet the TitanHQ channel team at the following MSP events to find out more about the TitanShield program for MSPs, OEMs, and service providers, and take advantage of the amazing new MSP package.

If you are unable to attend any of these events, be sure to give the TitanHQ team a call to find out more and take advantage of this exciting new and exclusive offer.
by titanadmin | Oct 21, 2019 | Email Scams, Phishing & Email Spam, Spam News, Spam Software |
A new Stripe phishing campaign has been detected that uses fake warnings advising users about an invalid account to lure people into divulging their credentials and bank account information.
Stripe is an online payment processor used by many online firms on their e-commerce websites to accept payments from their customers. As such, the company is perfect for spoofing as many people will be aware that the company processes payments and will think it reasonable that they need to provide credentials and bank account information to ensure payments are processed.
The scam starts with a phishing email supposedly from the Stripe Support department. The email advises the customer that the information associated with their account is currently invalid. The message is sent as a courtesy notice warning the user that their account will be placed on hold until the matter is corrected. The user is asked to review their details to correct the issue. A button is included in the email for users to click to do this.
The emails contain spelling mistakes and questionable grammar, so are likely to be identified as suspect by vigilant individuals. Security awareness training often teaches employees to hover their mouse arrow over a hyperlink to find out the true URL, but in this campaign it will not work. The attackers have added a title to the HTML tag of the embedded hyperlink so when the mouse arrow is hovered over the “Review your Details” button, that text will be displayed instead of the URL.
If that button is clicked, the user will be directed to a seemingly legitimate Stripe login page. The login box is a clone of the real login page and a series of boxes will be displayed, each requiring different information to be entered, including bank account and contact information.
When the user is required to enter their password, regardless of what is typed, the user will be advised that they have entered an incorrect password and will be asked to enter the password again. The user is then directed to the legitimate Stripe login page to make it appear they have been on the correct Stripe website all along.
Anti-Phishing Demo
Protect your MSP clients with the newest zero-day threat protection and intelligence against anti-phishing, business email compromise and zero-day attacks with PhishTitan.
Free Demo
Similar tactics are used in countless other phishing campaigns targeting other well-known companies. The presence of spelling mistakes and grammatical errors in messages should tip off end users that the email is a phishing attempt, but all too often end users fail to notice these errors and click and divulge sensitive information.
One issue is a lack of cybersecurity training in the workplace. If employees are not trained how to identify phishing emails, it is inevitable that some will end up falling for these scams and will divulge their credentials. Those credentials can be used to gain access to bank accounts or email accounts, with the latter often used to conduct further phishing attacks on the organization. One email account breach can easily lead to dozens of breached accounts.
For example, a phishing attack on a U.S. healthcare provider started with a single phishing email and led to 73 email accounts being compromised. As for cybersecurity awareness training, this is often nonexistent. One recent study on 2,000 employees in the United Kingdom revealed three quarters had received no workplace cybersecurity training whatsoever.
Protected by Microsoft Office 365 Anti-Phishing Controls? Are You Sure?
One in every 99 emails is a phishing email, so it is important to ensure your defenses are capable of blocking those messages. Many businesses mistakenly believe they are protected against these emails by Microsoft’s Office 365 anti-phishing controls. While those measures do block spam email and some phishing messages, one recent study by Avanan has shown 25% of phishing attacks sneak past Office 365 defenses and are delivered to inboxes. For an average firm that means several phishing emails will reach end users’ inboxes every day. To ensure your business is protected against phishing attacks, additional anti-phishing controls are required on top of Office 365.
Businesses can protect their Office 365 accounts against phishing by layering SpamTitan on top of Office 365. SpamTitan is an advanced anti-phishing and anti-malware solution that provides superior protection against phishing, malware, spear phishing, and zero-day attacks.
Heuristics rules are used to analyze message headers and these rules are constantly updated to include the latest threats. Bayesian analysis and heuristics are used to check message content, and along with machine learning techniques, new threats are blocked and prevented from reaching inboxes. Sandboxing is also used to assess email attachments for malicious code used to install malware in addition to dual-AV engines that scan for known malware.
These advanced measures ensure that Office 365 inboxes are kept free from malware and phishing emails. These advanced capabilities along with the ease of implementation and use and industry-leading customer support are why SpamTitan is the leading provider of anti-spam and anti-phishing solutions for SMBs and managed service providers that serve the SMB market.
For further information on SpamTitan, to book a product demonstration or set up a free trial, contact the TitanHQ team today.