Top 30 Critical Security Vulnerabilities Most Exploited by Hackers

CVE-2017-11882 is a memory corruption vulnerability (CWE-119) in Microsoft Office, residing in the legacy Microsoft Equation Editor component (EQNEDT32.EXE), that allows remote code execution in the context of the current user. Attackers trigger it by persuading a user to open a crafted document, most commonly an RTF file or other Office document carrying a malicious embedded equation object, which overflows a buffer while the equation content is parsed. Successful exploitation lets the attacker run arbitrary code with the privileges of the signed-in user, a typical foothold for malware delivery and, per CISA, for ransomware operations. Any environment running affected Microsoft Office builds is exposed; the source data does not enumerate specific affected version ranges. The flaw is confirmed exploited in the wild: it was added to the CISA Known Exploited Vulnerabilities catalog on 2021-11-03 with known ransomware use and holds a 99.9% EPSS score (percentile 100), though the source data lists no public PoC.

Do: Apply Microsoft's Office security updates (November 2017 or later) across all endpoints, prioritizing this KEV-listed flaw given its known ransomware use. On systems that cannot yet be patched, disable or unregister the legacy Equation Editor (EQNEDT32.EXE) and consider blocking or warning on RTF attachments as interim mitigations. Check for indicators of abuse such as EQNEDT32.EXE spawning unexpected child processes after document opens.

CVE-2018-13379 is a path traversal flaw (CWE-22) in the Fortinet FortiOS SSL VPN web portal that allows an unauthenticated attacker to download FortiOS system files via specially crafted HTTP resource requests. By traversing directories through crafted requests to the exposed web portal, the attacker can retrieve sensitive files, a technique publicly documented as yielding the SSL VPN session file containing usernames and passwords in plaintext. Any organization running the SSL VPN web portal on a FortiGate appliance is affected, and risk is highest where the portal is directly reachable from the internet. The flaw is confirmed in the wild: it was added to the CISA KEV catalog on 2021-11-03 with known ransomware use, and EPSS assigns it a 100% probability of exploitation within 30 days. No public PoC is listed in the provided data, but credential theft tied to this bug has been widely reused by threat actors.

Do: Apply the patched FortiOS release per Fortinet's vendor advisory immediately, as this is a CISA KEV required action; if the fixed version is not known from this data, follow Fortinet's FG-IR-18-384 advisory for the correct upgrade path. Rotate SSL VPN credentials and review VPN access logs for path-traversal requests, since successful exploitation exposes plaintext session credentials, and restrict SSL VPN portal exposure to trusted sources where possible.

CVE-2018-7600, widely known as 'Drupalgeddon 2', is an unauthenticated remote code execution flaw in Drupal Core caused by insufficient input validation (CWE-20) in how the CMS processes certain structured request data. It can be triggered through multiple attack vectors, such as crafted parameters submitted to commonly used form and rendering features that are reachable by anonymous users with a single HTTP request. Successful exploitation lets an attacker run arbitrary code under the web application, typically resulting in complete site compromise, and CISA notes the flaw has been used in ransomware operations. Any site running unpatched Drupal 7.x or 8.x core is affected; Drupal's installed base at the time of disclosure was on the order of one million sites. Exploitation is confirmed in the wild (CISA KEV, added 2021-11-03), EPSS assigns a 100% probability of exploitation within 30 days, and no public PoC is recorded in the supplied data.

Do: Upgrade immediately per vendor instructions: Drupal 7.58, 8.5.1, or the corresponding 8.4.6/8.3.9 updates if you remain on older 8.x branches, prioritizing internet-facing sites. Because this flaw has been exploited in the wild and used in ransomware operations, also check patched sites for backdoors, unexpected administrator accounts, modified core files, and rotate credentials.

Microsoft SharePoint fails to check the source markup of an application package, an improper input validation flaw (CWE-20) that allows maliciously crafted markup to be processed by the server. An attacker triggers the flaw by getting an affected SharePoint server to handle a crafted application package, without any special privileges described in the disclosure. Successful exploitation lets the attacker run remote code in the context of the SharePoint application pool and the SharePoint server farm account, providing control of the web server and access to a highly privileged farm-level identity. Any organization running an affected on-premises Microsoft SharePoint deployment is exposed, with internet-facing SharePoint servers at greatest risk. Exploitation is confirmed in the wild: the flaw was added to CISA's Known Exploited Vulnerabilities catalog on 2021-11-03 with known ransomware use and EPSS puts the probability of exploitation at 99.9%, although no public proof-of-concept is catalogued.

Do: Apply Microsoft's SharePoint security updates per vendor instructions immediately, prioritizing internet-exposed SharePoint servers as CISA's required action directs. Given known in-the-wild and ransomware use, hunt for signs of compromise such as unexpected .aspx or webshell files in SharePoint directories and anomalous use of the SharePoint farm account. Restrict or firewall internet exposure of SharePoint servers until patches are confirmed applied.

Ivanti Pulse Connect Secure, an enterprise SSL VPN appliance, contains an arbitrary file read vulnerability (CWE-22, path traversal) that requires no authentication. An unauthenticated remote attacker with network access to the appliance over HTTPS can send a specially crafted URI containing traversal sequences to read arbitrary files from the device. The attacker gains access to sensitive appliance files, potentially including configuration or credential material useful for further compromise, and CISA records known ransomware use of this flaw. Any organization running Pulse Connect Secure, especially gateways exposed to the internet for remote access, is affected. Exploitation is established: CISA added the flaw to its Known Exploited Vulnerabilities catalog on 2021-11-03 with known ransomware use, EPSS assigns a 100% probability of exploitation in the next 30 days, and no public PoC is known.

Do: Apply updates per Ivanti's instructions, the required action in the CISA KEV entry, prioritizing internet-facing Pulse Connect Secure gateways, and consult the vendor advisory for the applicable fixed release since no version range is provided here. Where patching is not immediate, restrict HTTPS access to the appliance and hunt for signs of exploitation (unexpected file reads, anomalous VPN logins or sessions, and follow-on ransomware activity), as CISA reports known ransomware use.

Atlassian accidentally shipped its pdkinstall development plugin enabled in release builds of Crowd and Crowd Data Center, so anyone who can send requests to the instance can install arbitrary plugins on it. An unauthenticated attacker sends crafted requests to a vulnerable Crowd server, uploads a malicious plugin, and gains remote code execution on the host. All releases from 2.1.0 through the 3.4.x line are affected until the 3.0.5, 3.1.6, 3.2.8, 3.3.5, and 3.4.4 fixes, so essentially every Crowd deployment predating those patches is exposed, particularly enterprises using Crowd for centralized identity/SSO with the instance reachable over the network. Exploitation is confirmed in the wild: the flaw is in CISA's Known Exploited Vulnerabilities catalog with known ransomware use, a public PoC exploit is available, and EPSS assigns a 95.4% probability of exploitation within 30 days.

Do: Upgrade Crowd or Crowd Data Center to 3.0.5, 3.1.6, 3.2.8, 3.3.5, or 3.4.4 (or later) on the corresponding release line; where patching is delayed, disable the pdkinstall plugin in the administration console as an interim mitigation. Review installed plugins for unfamiliar entries and hunt for web shells or post-exploitation activity, since CISA notes ransomware use of this flaw.

CVE-2019-18935 is a .NET deserialization flaw (CWE-502) in the RadAsyncUpload function of Progress Telerik UI for ASP.NET AJAX through version 2019.3.1023. It is triggered when an attacker who knows the Telerik upload encryption keys — most commonly because the earlier flaws CVE-2017-11317 or CVE-2017-11357 exposed them, though keys can be obtained by other means — sends crafted serialized data to RadAsyncUpload, allowing remote code execution without authentication. Successful exploitation gives an attacker arbitrary code execution on the hosting IIS/ASP.NET web server, reflected in the critical 9.8 CVSS score. Any web application built with Telerik UI for ASP.NET AJAX at or below 2019.3.1023 is affected, unless 2019.3.1023 has the non-default hardening setting enabled (as of 2020.1.114 a default setting prevents the exploit). Exploitation is rampant in the wild: the flaw was added to CISA KEV on 2021-11-03 with known ransomware use, carries a 99.7% EPSS probability of exploitation, has multiple public exploits (Bishop Fox, RAU_crypto, noperator), and has been used by multiple threat groups — including ransomware and government-linked actors — to breach organizations including a U.S. federal agency.

Do: Upgrade to Progress Telerik UI for ASP.NET AJAX 2020.1.114 or later, where a default setting prevents exploitation (or, if staying on 2019.3.1023, enable the non-default hardening setting); per CISA KEV, apply updates per vendor instructions. Because exploitation requires the encryption keys to be known, also patch the older CVE-2017-11317/CVE-2017-11357 key-disclosure flaws or rotate the Telerik upload encryption keys. Check internet-facing IIS/ASP.NET applications for exposed RadAsyncUpload handlers and indicators of compromise, given known ransomware and federal-agency breaches.

CVE-2019-19781 is a path-traversal flaw (classified CWE-22, though CISA's description calls it unspecified) in Citrix ADC (formerly NetScaler ADC), Citrix Gateway, and Citrix SD-WAN WANOP appliances that lets an unauthenticated remote attacker traverse directories via crafted requests and execute arbitrary commands on the appliance, typically with root privileges. It is triggered by sending specially crafted directory-traversal requests (crafted URLs/requests to the appliance's management or VPN endpoints), which lets the attacker write files and run commands with no credentials. Successful exploitation yields arbitrary code execution on the appliance, enabling theft of VPN/ADC credentials, lateral movement into the corporate network, and installation of persistent backdoors. Any organization running affected ADC, Gateway, or SD-WAN WANOP firmware is affected, with internet-facing gateways used for remote access at the highest risk. Exploitation is confirmed in the wild: the vulnerability is on CISA's KEV (added 2021-11-03) with known ransomware use, EPSS assigns near-certain (100.0%) probability of exploitation within 30 days, and no public PoC is listed despite confirmed abuse.

Do: Upgrade Citrix ADC, Gateway, and SD-WAN WANOP appliances to the fixed firmware builds listed in Citrix advisory CTX267020; if patching cannot be done immediately, apply Citrix's published interim mitigation and restrict internet exposure to the appliance. Because exploitation grants root code execution and persistence, after patching hunt for indicators of compromise (unexpected nsroot account, modified system files, crontab/scheduled entries), kill all active and inactive sessions, and rotate appliance and VPN credentials. Prioritize internet-facing gateways and comply with CISA's required action to apply vendor updates.

Atlassian Confluence Server and Data Center contain a server-side template injection flaw (CWE-22) in which attacker-supplied template content is processed by the server, enabling path traversal and ultimately remote code execution. An attacker triggers it by submitting crafted template syntax in a request to a vulnerable instance, and can run arbitrary commands or code on the underlying server. Successful compromise could allow data theft, deployment of webshells, or ransomware; CISA notes known ransomware use of this vulnerability. Organizations running self-hosted Confluence Server or Data Center are affected, particularly instances exposed to the internet. Exploitation is confirmed in the wild: the flaw is in CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03) and EPSS assigns a 99.9% probability of exploitation, placing it in the top percentile.

Do: Apply updates per vendor instructions by upgrading Confluence Server and Data Center to Atlassian's fixed releases, and prioritize internet-facing instances. Until patched, restrict network exposure of Confluence and monitor for signs of compromise such as webshells, unexpected processes, or ransomware activity, since ransomware operators are known to exploit this flaw.

CVE-2019-5591 is a default-configuration flaw in Fortinet FortiOS (CWE-306, missing authentication for a critical function): when a FortiGate acts as an LDAP client over SSL/TLS for authentication purposes, the default configuration does not properly authenticate the LDAP server, allowing an unauthenticated attacker on the same subnet to impersonate the LDAP server. The attack requires adjacency (the attacker must be on the same network segment as the device's LDAP traffic) and only works where the device uses LDAP over TLS with the default, non-validating settings. A successful attacker can intercept sensitive authentication information, such as usernames and passwords, exchanged with the LDAP server, with high confidentiality impact but no integrity or availability impact per the CVSS score. Organizations running affected FortiOS releases with LDAP-over-TLS configured are exposed; FortiOS runs on one of the largest firewall/VPN installed bases, though the source data does not state exact affected version ranges or deployment counts. The flaw is listed in CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03) with known in-the-wild and ransomware use; late-2021 CISA and FBI advisories warned that APT actors were exploiting Fortinet bugs, including this one, for initial access, and no public PoC is known.

Do: Upgrade affected FortiGate/FortiOS units to a fixed release (6.2.1+, 6.0.8+, 5.6.10+, or 5.4.12+) per Fortinet's instructions, as required by CISA's KEV listing. If patching is delayed, disable the use of SSL/TLS in the LDAP server configuration or restrict and monitor the network path between the FortiGate and the LDAP server. Inventory which devices use LDAPS for authentication and review authentication logs for signs of server impersonation or credential theft.

CVE-2020-0688 is a remote code execution vulnerability in Microsoft Exchange Server caused by the validation key not being uniquely created at install time, leaving deployments with a predictable, shared key (CWE-287, improper authentication). A remote attacker who can reach an affected Exchange server and knows the common install-time key can supply maliciously crafted, cryptographically signed payloads that the server trusts, triggering code execution without needing per-server secrets. Successful exploitation gives the attacker code execution on the Exchange server, which can be used to access mail data, move laterally, and stage follow-on activity; CISA notes known use in ransomware campaigns. All organizations running the affected on-premises Microsoft Exchange Server are in scope per CISA's listing, though the affected version range is not specified in the source data. Exploitation is confirmed in the wild: the flaw was added to CISA's KEV on 2021-11-03 with known ransomware use, and EPSS rates 30-day exploitation probability at 100% (top percentile).

Do: Apply Microsoft's Exchange security updates addressing CVE-2020-0688 (released in February 2020) to every on-premises Exchange server, per CISA's required action. As an interim mitigation, configure a unique ASP.NET machineKey in each Exchange server's web.config instead of the default shared install-time key, and hunt for indicators of exploitation given the known ransomware use.

CVE-2020-0787 is a privilege elevation flaw in the Windows Background Intelligent Transfer Service (BITS), which improperly handles symbolic links (CWE-269, CWE-59). An attacker who can already run low-privileged code on a machine — via a phishing payload or a chained remote-code-execution bug — can plant or manipulate symbolic links that BITS follows, causing the service to execute arbitrary code with SYSTEM-level privileges. Successful exploitation grants full control of the host, making this a common link in attack chains, and CISA notes known ransomware use. Because BITS ships by default with Windows, essentially every Windows client and server installation predating the vendor patch is affected. Exploitation is confirmed in the wild: the flaw was added to CISA's Known Exploited Vulnerabilities catalog on 2022-01-28, and EPSS assigns a 42.5% probability of exploitation within 30 days (99th percentile).

Do: Apply Microsoft's updates for CVE-2020-0787 per vendor instructions (the flaw was addressed in Microsoft's March 2020 security updates) across all Windows clients and servers, prioritizing user workstations and internet-facing systems given known ransomware use. Since this is a local privilege escalation, pair patching with controls that block the initial foothold (MFA, email/phishing defenses, EDR). Verify remediation by confirming endpoints report the relevant update installed and no BITS symlink abuse indicators remain.

Fortinet FortiOS SSL VPN improperly validates the second factor of authentication (FortiToken), allowing a user with valid primary credentials to log in without being prompted for their one-time token. The bypass is triggered by changing the letter case of the username (e.g., typing 'Admin' instead of 'admin'), a case-sensitivity handling flaw (CWE-178) in the authentication logic (CWE-287). An attacker who has stolen, phished, or reused a user's primary credentials can therefore gain full SSL VPN access without the multi-factor protection defenders assume is enforced. Any organization running an affected FortiOS SSL VPN portal with FortiToken two-factor authentication for users is exposed. The flaw is in CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03) with known ransomware use, and EPSS assigns a 49.3% probability of exploitation within 30 days (99th percentile); no public PoC is known.

Do: Apply the FortiOS update per Fortinet's advisory, as required by CISA's KEV listing. Until patched, audit SSL VPN logs for successful logins in which no FortiToken second-factor prompt occurred (especially sessions authenticating with altered username case), and consider restricting the SSL VPN portal to trusted source addresses or disabling it temporarily. Any accounts found with second-factor-free VPN logins should have credentials rotated and be checked for follow-on compromise, given known ransomware use.

CVE-2020-1472, widely known as "Zerologon," is an elevation-of-privilege flaw in how the Netlogon secure channel is established over the Netlogon Remote Protocol (MS-NRPC) on Microsoft domain controllers. An unauthenticated attacker with network reachability to a domain controller sends specially crafted Netlogon messages to establish a vulnerable secure channel and then runs a specially crafted application on the network to obtain domain administrator access. Successful exploitation yields domain administrator privileges, effectively full compromise of the Active Directory environment, and the flaw is known to be used in ransomware operations. Any organization running affected Windows Server versions (2008 through 20H2) as domain controllers is exposed, along with environments using Netlogon implementations from Samba and distributions or products from Fedora, openSUSE, Canonical (Ubuntu), Debian, Synology, and Oracle. Exploitation is highly active: a public Zerologon PoC/exploit is available, the flaw is on CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03, with known ransomware use), and EPSS estimates a 99.4% probability of exploitation within 30 days.

Do: Apply the vendor updates on all domain controllers and other affected systems immediately, following Microsoft's two-phase Netlogon secure channel guidance (the enforcement phase of the phased rollout began in Q1 2021). Audit Netlogon secure-channel connections and event logs for clients still using vulnerable connections before enabling full enforcement, and install updated packages for Samba and other Netlogon implementations from Fedora, openSUSE, Ubuntu, Debian, Synology, and Oracle. Given known ransomware use, prioritize patching any domain controller reachable from user networks, VPNs, or the internet.

CVE-2020-15505 is a critical, unauthenticated remote code execution vulnerability in Ivanti MobileIron's Core and Enterprise Connector, Sentry, and Monitor and Reporting Database (RDB) products, which public proof-of-concept exploits identify as a Hessian-based Java deserialization flaw. An unauthenticated attacker can send crafted requests to an affected MobileIron server over the network, with no privileges or user interaction required, and execute arbitrary code, gaining full control of the MDM server with high impact on confidentiality, integrity, and availability. Any organization running the affected versions of these enterprise mobile device management products is at risk, particularly internet-facing MobileIron Core and Sentry instances. Exploitation is confirmed and widespread: the flaw is on CISA's Known Exploited Vulnerabilities catalog (added November 3, 2021), carries a 99.7% EPSS score, and was named among the top flaws exploited by Chinese state-sponsored hackers, prompting urgent UK NCSC patching alerts. It is also being observed chained with other exploited vulnerabilities (e.g., VPN flaws and Zerologon) as an initial-access vector, so defenders should treat it as actively exploited.

Do: Apply the vendor updates immediately per Ivanti's security advisory for MobileIron Core, Enterprise Connector, Sentry, and RDB (this is CISA's required action for KEV entries). Until patched, reduce internet exposure of MobileIron interfaces and monitor for exploitation; because the flaw was mass-exploited (including by Chinese state-sponsored actors), hunt for signs of compromise such as unexpected processes, persistence mechanisms, or webshells on affected MDM servers.

CVE-2020-5902 is a critical, unauthenticated remote code execution flaw in the F5 BIG-IP Traffic Management User Interface (TMUI), the appliance's web management console, rooted in a directory/path traversal issue (CWE-22) in undisclosed TMUI pages. It is triggered by sending crafted HTTP(S) requests to the management interface — classically path-traversal URLs beneath the TMUI application on the management port — which lets an attacker bypass authentication, read or delete arbitrary files, and execute commands without credentials. Successful exploitation yields full control of the BIG-IP system, which attackers can use to pivot into networks the appliance fronts, maintain persistence, and deploy ransomware. Any organization running an affected F5 BIG-IP appliance or virtual edition whose TMUI is reachable, or whose management network can be reached, is exposed; F5's installed base spans large enterprises and service providers, so the footprint is broad. Exploitation is confirmed in the wild: the flaw was mass-scanned and exploited within days of its July 2020 disclosure, it is listed in CISA KEV with known ransomware use, and EPSS assigns a ~100% probability of exploitation within 30 days.

Do: Patch immediately using F5's advisory K52145254 — upgrade BIG-IP to a fixed release per the vendor's version matrix, since CISA's required action is applying vendor updates. Until patched, restrict TMUI/management-interface access to trusted source IPs or a VPN (or disable TMUI if unused) and apply F5's published interim workaround. Because ransomware use is confirmed, hunt for indicators of compromise on both patched and unpatched appliances (unexpected files, webshells, modified login pages, new accounts or scheduled tasks) before treating systems as clean.

CVE-2021-21985 is an improper input validation flaw (CWE-20, with related unsafe reflection CWE-470 and SSRF CWE-918 classifications) in the Virtual SAN Health Check plug-in of the VMware vSphere Client, which is enabled by default in vCenter Server. It is triggered by crafted requests sent to the plug-in over the network; VMware indicated that network access to vCenter's HTTPS port (443) is sufficient to reach the vulnerable component. A successful attacker gains remote code execution with unrestricted privileges on the underlying operating system hosting vCenter Server, a highly privileged position in the virtualization stack. Any organization running an affected VMware vCenter Server is affected; because vCenter is the default management plane for vSphere, this spans a very large share of enterprise virtualization estates, with tens of thousands of instances directly exposed to the internet. Exploitation is confirmed in the wild: CISA added the flaw to the Known Exploited Vulnerabilities catalog on 2021-11-03 with known ransomware use and requires applying vendor updates, and EPSS puts the 30-day exploitation probability at essentially 100%, though the source data lists no public PoC.

Do: Update vCenter Server per VMware's instructions; fixes shipped in May 2021 for the 6.5, 6.7, and 7.0 branches (e.g., 6.5 U3n, 6.7 U3o, and 7.0 U2c — verify your current build against the vendor advisory). Until patched, restrict access to vCenter's HTTPS (443) interface to trusted management networks rather than the open internet, and review appliance logs and running processes for indicators of exploitation, since ransomware operators are known to use this flaw after gaining network access.

Ivanti Pulse Connect Secure, a widely deployed SSL VPN appliance, contains a use-after-free vulnerability in its license services. A remote, unauthenticated attacker can trigger the flaw via the license services and gain arbitrary code execution on the appliance, which is a high-value target because it terminates VPN sessions for enterprise networks. Any organization running an affected Pulse Connect Secure release is potentially affected; the source data does not specify exact version ranges, so administrators should compare their release against Ivanti's advisory. Exploitation is confirmed in the wild: the flaw is on CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03) with known ransomware use, and its 47.2% EPSS score (99th percentile) indicates a high likelihood of exploitation within 30 days.

Do: Apply the updates per Ivanti's instructions immediately, as CISA's required action specifies. Because exploited appliances have often retained persistent webshells/backdoors even after patching, also hunt for indicators of compromise (modified appliance files, unexpected processes or accounts) and follow Ivanti's remediation guidance rather than only installing the update. Until patched, restrict or closely monitor internet access to the appliance.

CVE-2021-22894 is a buffer overflow in the Collaboration Suite (meeting room) functionality of Ivanti Pulse Connect Secure, classified under CWE-94 (improper control of code generation). A remote attacker who is already authenticated to the appliance can trigger the overflow by supplying a maliciously crafted meeting room, causing attacker-controlled code to run on the appliance. Successful exploitation yields code execution as the root user, giving the attacker full control of the VPN appliance at the operating-system level. Any organization running Pulse Connect Secure with the Collaboration Suite feature enabled is potentially affected; specific affected version ranges are not specified in the available data. The flaw is on CISA's Known Exploited Vulnerabilities catalog (added 2021-11-03), confirming in-the-wild exploitation, with an elevated 41.3% EPSS probability of exploitation in the next 30 days and no public proof-of-concept known.

Do: Apply Ivanti's supplied updates for Pulse Connect Secure per vendor instructions immediately, as the flaw is on CISA's KEV list with a required-action deadline. Restrict or disable the Collaboration/meeting feature if not needed, ensure it is not reachable without authentication, and review appliance and authentication logs for signs of exploitation (unexpected processes, files, or sessions running as root).

CVE-2021-26855 is a server-side request forgery flaw (CWE-918) in Microsoft Exchange Server that allows an unauthenticated remote attacker to send specially crafted HTTP requests and have the Exchange server process them as itself, disclosing sensitive session information. When chained with sibling Exchange flaws (the 'ProxyLogon' chain), it yields authentication bypass and arbitrary file write, escalating to full remote code execution with SYSTEM-level privileges on the on-premises Exchange server. Any organization running an affected on-premises Exchange server reachable over HTTP/HTTPS (typically outbound webmail) is exposed; Exchange Online was not affected. Exploitation is confirmed in the wild at large scale: the flaw was mass-exploited beginning in early 2021 (notably by the HAFNIUM group), is on the CISA KEV with documented ransomware use, and has a maximum EPSS score of 100% (100th percentile), despite no public PoC listing.

Do: Apply the vendor's March 2021 Exchange security updates (or later cumulative updates) immediately, per the CISA required action; until patched, limit Exchange (ECP/OWA) exposure to the internet via firewall/VPN rules. Hunt for compromise: review IIS logs for unrecognized authenticated activity against FrontEnd HttpProxy endpoints, and check for malicious files or webshells under inetpub\wwwroot\aspnet_client, given the known ransomware use.

CVE-2021-27104 is an OS command injection flaw (CWE-78) in Accellion's File Transfer Appliance (FTA), affecting versions 9_12_370 and earlier. It is triggered by sending a crafted POST request to various admin endpoints, and the CVSS vector (no privileges, no user interaction, network-accessible) indicates it can be exploited by an unauthenticated remote attacker. Successful exploitation yields full OS command execution on the appliance, giving the attacker control sufficient for data theft, web shell deployment, and follow-on ransomware/extortion operations. Organizations running Accellion FTA appliances — typically deployed as internet-facing large-file transfer endpoints by enterprises, government agencies, and universities — are affected. The flaw is being actively exploited in the wild: it was added to CISA's KEV catalog on 2021-11-03 with known ransomware use (notably the Clop/FINEST data-theft extortion campaign, including the breach of security firm Qualys), and it carries a high EPSS score of 56.7% (99th percentile). The fixed version is FTA_9_12_380 and later.

Do: Upgrade Accellion FTA to version FTA_9_12_380 or later per vendor instructions; the flaw is on CISA's KEV list with known ransomware use, so patching is urgent for internet-facing appliances. Until patched, restrict or firewall access to FTA admin endpoints from the internet, and review logs and the appliance for signs of command injection or web shell (e.g., DEWMODE-related) compromise given the active data-theft extortion campaign.