PulsesecureCVEs & Vulnerabilities

Jonathan Looney discovered that the TCP retransmission queue implementation in tcp_fragment in the Linux kernel could be fragmented when handling certain TCP Selective Acknowledgment (SACK) sequences. A remote attacker could use this to cause a denial of service. This has been fixed in stable kernel releases 4.4.182, 4.9.182, 4.14.127, 4.19.52, 5.1.11, and is fixed in commit f070ef2ac66716357066b683fb0baf55f8191a2e.

Jonathan Looney discovered that the TCP_SKB_CB(skb)->tcp_gso_segs value was subject to an integer overflow in the Linux kernel when handling TCP Selective Acknowledgments (SACKs). A remote attacker could use this to cause a denial of service. This has been fixed in stable kernel releases 4.4.182, 4.9.182, 4.14.127, 4.19.52, 5.1.11, and is fixed in commit 3b4929f65b0d8249f19a50245cd88ed1a2f78cff.

In Pulse Secure Pulse Connect Secure (PCS) before 8.1R15.1, 8.2 before 8.2R12.1, 8.3 before 8.3R7.1, and 9.0 before 9.0R3.4 and Pulse Policy Secure (PPS) before 5.1R15.1, 5.2 before 5.2R12.1, 5.3 before 5.3R15.1, 5.4 before 5.4R7.1, and 9.0 before 9.0R3.2, an authenticated attacker (via the admin web interface) can exploit Incorrect Access Control to execute arbitrary code on the appliance.

In Pulse Secure Pulse Connect Secure (PCS) before 8.1R15.1, 8.2 before 8.2R12.1, 8.3 before 8.3R7.1, and 9.0 before 9.0R3.4, an authenticated attacker (via the admin web interface) can exploit Directory Traversal to execute arbitrary code on the appliance.

XSS exists in the admin web console in Pulse Secure Pulse Connect Secure (PCS) 9.0RX before 9.0R3.4, 8.3RX before 8.3R7.1, and 8.1RX before 8.1R15.1 and Pulse Policy Secure 9.0RX before 9.0R3.2, 5.4RX before 5.4R7.1, and 5.2RX before 5.2R12.1.

In Pulse Secure Pulse Connect Secure version 9.0RX before 9.0R3.4, 8.3RX before 8.3R7.1, 8.2RX before 8.2R12.1, and 8.1RX before 8.1R15.1 and Pulse Policy Secure version 9.0RX before 9.0R3.2, 5.4RX before 5.4R7.1, 5.3RX before 5.3R12.1, 5.2RX before 5.2R12.1, and 5.1RX before 5.1R15.1, an authenticated attacker (via the admin web interface) can send a specially crafted message resulting in a stack buffer overflow.

In Pulse Secure Pulse Connect Secure version 9.0RX before 9.0R3.4, 8.3RX before 8.3R7.1, and 8.2RX before 8.2R12.1, users using SAML authentication with the Reuse Existing NC (Pulse) Session option may see authentication leaks.

In Pulse Secure Pulse Connect Secure version 9.0RX before 9.0R3.4 and 8.3RX before 8.3R7.1 and Pulse Policy Secure version 9.0RX before 9.0R3.2 and 5.4RX before 5.4R7.1, an unauthenticated, remote attacker can conduct a session hijacking attack.

In Pulse Secure Pulse Desktop Client and Network Connect, an attacker could access session tokens to replay and spoof sessions, and as a result, gain unauthorized access as an end user, a related issue to CVE-2019-1573. (The endpoint would need to be already compromised for exploitation to succeed.) This affects Pulse Desktop Client 5.x before Secure Desktop 5.3R7 and Pulse Desktop Client 9.x before Secure Desktop 9.0R3. It also affects (for Network Connect customers) Pulse Connect Secure 8.1 before 8.1R14, 8.3 before 8.3R7, and 9.0 before 9.0R3.

Certain Secure Access SA Series SSL VPN products (originally developed by Juniper Networks but now sold and supported by Pulse Secure, LLC) allow privilege escalation, as demonstrated by Secure Access SSL VPN SA-4000 5.1R5 (build 9627) 4.2 Release (build 7631). This occurs because appropriate controls are not performed. Specifically, it is possible for a readonly user to change the administrator user password by making a local copy of the /dana-admin/user/update.cgi page, changing the "user" value, and saving the changes.

Pulse Secure Virtual Traffic Manager 9.9 versions prior to 9.9r2 and 10.4r1 allow a remote authenticated user to obtain sensitive historical activity information by leveraging incorrect permission validation.

A stored cross-site scripting (XSS) vulnerability in the web administration user interface of Pulse Secure Virtual Traffic Manager may allow a remote authenticated attacker to inject web script or HTML via a crafted website and steal sensitive data and credentials. Affected releases are Pulse Secure Virtual Traffic Manager 9.9 versions prior to 9.9r2 and 10.4r1.

Pulse Secure Desktop Client 5.3 up to and including R6.0 build 1769 on Windows has Insecure Permissions.

Artifex Ghostscript 9.25 and earlier allows attackers to bypass a sandbox protection mechanism via vectors involving the 1Policy operator.

Pulse Secure Client 9.0R1 and 5.3RX before 5.3R5, when configured to authenticate VPN users during Windows Logon, can allow attackers to bypass Windows authentication and execute commands on the system with the privileges of Pulse Secure Client. The attacker must interrupt the client's network connectivity, and trigger a connection to a crafted proxy server with an invalid SSL certificate that allows certification-manager access, leading to the ability to browse local files and execute local programs.

A vulnerability has been discovered in login.cgi in Pulse Secure Pulse Connect Secure (PCS) 8.1RX before 8.1R12 and 8.3RX before 8.3R2 and Pulse Policy Secure (PPS) 5.2RX before 5.2R9 and 5.4RX before 5.4R2 wherein an http(s) Host header received from the browser is trusted without validation.

In Pulse Secure Pulse Desktop Client 5.3RX before 5.3R5 and 9.0R1, there is a Privilege Escalation Vulnerability with Dynamic Certificate Trust.

The Pulse Secure Desktop (macOS) has a Privilege Escalation Vulnerability.

The Pulse Secure Desktop (macOS) 5.3RX before 5.3R5 and 9.0R1 has a Format String Vulnerability.

The Pulse Secure Desktop (macOS) 5.3RX before 5.3R5 and 9.0R1 has a Privilege Escalation Vulnerability.

download.cgi in Pulse Secure Pulse Connect Secure 8.1RX before 8.1R13 and 8.3RX before 8.3R4 and Pulse Policy Secure through 5.2RX before 5.2R10 and 5.4RX before 5.4R4 have an Open Redirect Vulnerability.

In Artifex Ghostscript before 9.24, attackers able to supply crafted PostScript files could use a type confusion in the setcolor function to crash the interpreter or possibly have unspecified other impact.

In Artifex Ghostscript 9.23 before 2018-08-24, attackers able to supply crafted PostScript could use uninitialized memory access in the aesdecode operator to crash the interpreter or potentially execute code.

In Artifex Ghostscript before 9.24, attackers able to supply crafted PostScript files could use a type confusion in the LockDistillerParams parameter to crash the interpreter or execute code.

In Artifex Ghostscript 9.23 before 2018-08-24, a type confusion using the .shfill operator could be used by attackers able to supply crafted PostScript files to crash the interpreter or potentially execute code.

Pulse Secure Pulse Connect Secure 8.1.x before 8.1R14, 8.2.x before 8.2R11, and 8.3.x before 8.3R5 do not properly process nested XML entities, which allows remote attackers to cause a denial of service (memory consumption and memory errors) via a crafted XML document.

The GUI component (aka PulseUI) in Pulse Secure Desktop Linux clients before PULSE5.2R9.2 and 5.3.x before PULSE5.3R4.2 does not perform strict SSL Certificate Validation. This can lead to the manipulation of the Pulse Connection set.

A stack-based Buffer Overflow Vulnerability exists in the web server in Pulse Secure Pulse Connect Secure (PCS) before 8.3R4 and Pulse Policy Secure (PPS) before 5.4R4, leading to memory corruption and possibly remote code execution.

A cross site scripting issue has been found in custompage.cgi in Pulse Secure Pulse Connect Secure (PCS) before 8.0R17.0, 8.1.x before 8.1R13, 8.2.x before 8.2R9, and 8.3.x before 8.3R3 and Pulse Policy Secure (PPS) before 5.2R10, 5.3.x before 5.3R9, and 5.4.x before 5.4R3 due to one of the URL parameters not being sanitized. Exploitation does require the user to be logged in as administrator; the issue is not applicable to the end user portal.

Pulse Secure Pulse One On-Premise 2.0.1649 and below does not properly validate requests, which allows remote users to query and obtain sensitive information.

diag.cgi in Pulse Connect Secure 8.2R1 through 8.2R5, 8.1R1 through 8.1R10 and Pulse Policy Secure 5.3R1 through 5.3R5, 5.2R1 through 5.2R8, and 5.1R1 through 5.1R10 allow remote attackers to hijack the authentication of administrators for requests to start tcpdump, related to the lack of anti-CSRF tokens.

Pulse Connect Secure 8.3R1 has CSRF in logout.cgi. The logout function of the admin panel is not protected by any CSRF tokens, thus allowing an attacker to logout a user by making them visit a malicious web page.

Pulse Connect Secure 8.3R1 has Reflected XSS in launchHelp.cgi. The helpLaunchPage parameter is reflected in an IFRAME element, if the value contains two quotes. It properly sanitizes quotes and tags, so one cannot simply close the src with a quote and inject after that. However, an attacker can use javascript: or data: to abuse this.

Pulse Connect Secure 8.3R1 has Reflected XSS in adminservercacertdetails.cgi. In the admin panel, the certid parameter of adminservercacertdetails.cgi is reflected in the application's response and is not properly sanitized, allowing an attacker to inject tags. An attacker could come up with clever payloads to make the system run commands such as ping, ping6, traceroute, nslookup, arp, etc.

Pulse Connect Secure 8.3R1 has CSRF in diag.cgi. In the panel, the diag.cgi file is responsible for running commands such as ping, ping6, traceroute, traceroute6, nslookup, arp, and Portprobe. These functions do not have any protections against CSRF. That can allow an attacker to run these commands against any IP if they can get an admin to visit their malicious CSRF page.

Pulse Secure Desktop before 5.2R2 and Pulse Secure Installer Service before 8.2R2 and below for Windows allow restricted users to gain privileges via unspecified vectors.

The administrative user interface in Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r2, 8.0 before 8.0r9, and 7.4 before 7.4r13.4 allows remote administrators to enumerate files, read arbitrary files, and conduct server side request forgery (SSRF) attacks via unspecified vectors.

Cross-site scripting (XSS) vulnerability in the administrative user interface in Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r2, 8.0 before 8.0r9, and 7.4 before 7.4r13.4 allows remote attackers to inject arbitrary web script or HTML via unspecified vectors.

Cross-site scripting (XSS) vulnerability in the system configuration section in the administrative user interface in Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r2, 8.0 before 8.0r9, and 7.4 before 7.4r13.4 allows remote attackers to inject arbitrary web script or HTML via unspecified vectors.

Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r2, 8.0 before 8.0r10, and 7.4 before 7.4r13.4 allow remote attackers to read an unspecified system file via unknown vectors.

Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r2, 8.0 before 8.0r10, and 7.4 before 7.4r13.4 allow remote attackers to read sensitive system authentication files in an unspecified directory via unknown vectors.

Pulse Connect Secure (PCS) 8.2 before 8.2r1, 8.1 before 8.1r3, 8.0 before 8.0r11, and 7.4 before 7.4r13.4 allow remote attackers to cause a denial of service (CPU consumption) via unspecified vectors.

The Terminal Services Remote Desktop Protocol (RDP) client session restrictions feature in Pulse Connect Secure (aka PCS) 8.1R7 and 8.2R1 allow remote authenticated users to bypass intended access restrictions via unspecified vectors.

The fmtstr function in crypto/bio/b_print.c in OpenSSL 1.0.1 before 1.0.1s and 1.0.2 before 1.0.2g improperly calculates string lengths, which allows remote attackers to cause a denial of service (overflow and out-of-bounds read) or possibly have unspecified other impact via a long string, as demonstrated by a large amount of ASN.1 data, a different vulnerability than CVE-2016-2842.

The SSLv2 protocol, as used in OpenSSL before 1.0.1s and 1.0.2 before 1.0.2g and other products, requires a server to send a ServerVerify message before establishing that a client possesses certain plaintext RSA data, which makes it easier for remote attackers to decrypt TLS ciphertext data by leveraging a Bleichenbacher RSA padding oracle, aka a "DROWN" attack.