When using the Apache JServ Protocol (AJP), care must be taken when trusting incoming connections to Apache Tomcat. Tomcat treats AJP connections as having higher trust than, for example, a similar HTTP connection. If such connections are available to an attacker, they can be exploited in ways that may be surprising. In Apache Tomcat 9.0.0.M1 to 9.0.0.30, 8.5.0 to 8.5.50 and 7.0.0 to 7.0.99, Tomcat shipped with an AJP Connector enabled by default that listened on all configured IP addresses. It was expected (and recommended in the security guide) that this Connector would be disabled if not required. This vulnerability report identified a mechanism that allowed: - returning arbitrary files from anywhere in the web application - processing any file in the web application as a JSP Further, if the web application allowed file upload and stored those files within the web application (or the attacker was able to control the content of the web application by some other means) then this, along with the ability to process a file as a JSP, made remote code execution possible. It is important to note that mitigation is only required if an AJP port is accessible to untrusted users. Users wishing to take a defence-in-depth approach and block the vector that permits returning arbitrary files and execution as JSP may upgrade to Apache Tomcat 9.0.31, 8.5.51 or 7.0.100 or later. A number of changes were made to the default AJP Connector configuration in 9.0.31 to harden the default configuration. It is likely that users upgrading to 9.0.31, 8.5.51 or 7.0.100 or later will need to make small changes to their configurations.

scm/define-stencil-commands.scm in LilyPond through 2.20.0, and 2.21.x through 2.21.4, when -dsafe is used, lacks restrictions on embedded-ps and embedded-svg, as demonstrated by including dangerous PostScript code.

SQL injection vulnerability in OpenVAS Manager before 4.0.6 and 5.x before 5.0.7 allows remote attackers to execute arbitrary SQL commands via the timezone parameter in a modify_schedule OMP command.

The tt_cmap4_validate function in sfnt/ttcmap.c in FreeType before 2.5.4 validates a certain length field before that field's value is completely calculated, which allows remote attackers to cause a denial of service (out-of-bounds read) or possibly have unspecified other impact via a crafted cmap SFNT table.

default-rsyncssh.lua in Lsyncd 2.1.5 and earlier allows remote attackers to execute arbitrary commands via shell metacharacters in a filename.

The tt_face_load_hdmx function in truetype/ttpload.c in FreeType before 2.5.4 does not establish a minimum record size, which allows remote attackers to cause a denial of service (out-of-bounds read) or possibly have unspecified other impact via a crafted TrueType font.

LibreOffice before 4.3.5 allows remote attackers to cause a denial of service (invalid write operation and crash) and possibly execute arbitrary code via a crafted RTF file.

cff/cf2ft.c in FreeType before 2.5.4 does not validate the return values of point-allocation functions, which allows remote attackers to cause a denial of service (heap-based buffer overflow) or possibly have unspecified other impact via a crafted OTF font.

The _bdf_parse_glyphs function in bdf/bdflib.c in FreeType before 2.5.4 does not properly handle a missing ENDCHAR record, which allows remote attackers to cause a denial of service (NULL pointer dereference) or possibly have unspecified other impact via a crafted BDF font.

The _bfd_XXi_swap_aouthdr_in function in bfd/peXXigen.c in GNU binutils 2.24 and earlier allows remote attackers to cause a denial of service (out-of-bounds write) and possibly have other unspecified impact via a crafted NumberOfRvaAndSizes field in the AOUT header in a PE executable.

An issue was discovered in adns before 1.5.2. It hangs, eating CPU, if a compression pointer loop is encountered.

The setup_group function in bfd/elf.c in libbfd in GNU binutils 2.24 and earlier allows remote attackers to cause a denial of service (crash) and possibly execute arbitrary code via crafted section group headers in an ELF file.

pam_ldap in nss_ldap on Red Hat Enterprise Linux 4, Fedora Core 3 and earlier, and possibly other distributions does not return an error condition when an LDAP directory server responds with a PasswordPolicyResponse control response, which causes the pam_authenticate function to return a success code even if authentication has failed, as originally reported for xscreensaver.

Stack-based buffer overflow in the ihex_scan function in bfd/ihex.c in GNU binutils 2.24 and earlier allows remote attackers to cause a denial of service (crash) and possibly have other unspecified impact via a crafted ihex file.

Stack-based buffer overflow in the srec_scan function in bfd/srec.c in GNU binutils 2.24 and earlier allows remote attackers to cause a denial of service (crash) and possibly have other unspecified impact via a crafted file.

Heap-based buffer overflow in the pe_print_edata function in bfd/peXXigen.c in GNU binutils 2.24 and earlier allows remote attackers to cause a denial of service (crash) and possibly have other unspecified impact via a truncated export table in a PE file.

SQL injection vulnerability in Zend Framework before 1.12.9, 2.2.x before 2.2.8, and 2.3.x before 2.3.3, when using the sqlsrv PHP extension, allows remote attackers to execute arbitrary SQL commands via a null byte.

389-ds-base version before 1.3.5.19 and 1.3.6.7 are vulnerable to password brute-force attacks during account lockout due to different return codes returned on password attempts.

Integer overflow in the MallocFrameBuffer function in vncviewer.c in LibVNCServer 0.9.9 and earlier allows remote VNC servers to cause a denial of service (crash) and possibly execute arbitrary code via an advertisement for a large screen size, which triggers a heap-based buffer overflow.

visionmedia send before 0.8.4 for Node.js uses a partial comparison for verifying whether a directory is within the document root, which allows remote attackers to access restricted directories, as demonstrated using "public-restricted" under a "public" directory.

A URL parameter injection vulnerability was found in the back-channel ticket validation step of the CAS protocol in Jasig Java CAS Client before 3.3.2, .NET CAS Client before 1.0.2, and phpCAS before 1.3.3 that allow remote attackers to inject arbitrary web script or HTML via the (1) service parameter to validation/AbstractUrlBasedTicketValidator.java or (2) pgtUrl parameter to validation/Cas20ServiceTicketValidator.java.

An issue was discovered in SaltStack Salt through 3002. Sending crafted web requests to the Salt API, with the SSH client enabled, can result in shell injection.

XML external entity (XXE) vulnerability in Zabbix 1.8.x before 1.8.21rc1, 2.0.x before 2.0.13rc1, 2.2.x before 2.2.5rc1, and 2.3.x before 2.3.2 allows remote attackers to read arbitrary files or potentially execute arbitrary code via a crafted DTD in an XML request.

In the standard library in Rust before 1.52.0, a double free can occur in the Vec::from_iter function if freeing the element panics.

The nsXBLProtoImpl::InstallImplementation function in Mozilla Firefox before 29.0, Firefox ESR 24.x before 24.5, Thunderbird before 24.5, and SeaMonkey before 2.26 does not properly check whether objects are XBL objects, which allows remote attackers to execute arbitrary code or cause a denial of service (buffer overflow) via crafted JavaScript code that accesses a non-XBL object as if it were an XBL object.

Use-after-free vulnerability in the nsHostResolver::ConditionallyRefreshRecord function in libxul.so in Mozilla Firefox before 29.0, Firefox ESR 24.x before 24.5, Thunderbird before 24.5, and SeaMonkey before 2.26 allows remote attackers to execute arbitrary code or cause a denial of service (heap memory corruption) via vectors related to host resolution.

common/session.c in Claws Mail before 3.17.6 has a protocol violation because suffix data after STARTTLS is mishandled.

It was discovered that websockets.c in LibVNCServer prior to 0.9.12 did not properly decode certain WebSocket frames. A malicious attacker could exploit this by sending specially crafted WebSocket frames to a server, causing a heap-based buffer overflow.

Integer underflow in the LCodeGen::PrepareKeyedOperand function in arm/lithium-codegen-arm.cc in Google V8 before 3.25.28.16, as used in Google Chrome before 35.0.1916.114, allows remote attackers to cause a denial of service or possibly have unspecified other impact via vectors that trigger a negative key value.

In Symfony before versions 4.4.13 and 5.1.5, the CachingHttpClient class from the HttpClient Symfony component relies on the HttpCache class to handle requests. HttpCache uses internal headers like X-Body-Eval and X-Body-File to control the restoration of cached responses. The class was initially written with surrogate caching and ESI support in mind (all HTTP calls come from a trusted backend in that scenario). But when used by CachingHttpClient and if an attacker can control the response for a request being made by the CachingHttpClient, remote code execution is possible. This has been fixed in versions 4.4.13 and 5.1.5.

The bubblewrap sandbox of WebKitGTK and WPE WebKit, prior to 2.28.3, failed to properly block access to CLONE_NEWUSER and the TIOCSTI ioctl. CLONE_NEWUSER could potentially be used to confuse xdg-desktop-portal, which allows access outside the sandbox. TIOCSTI can be used to directly execute commands outside the sandbox by writing to the controlling terminal's input buffer, similar to CVE-2017-5226.

ClamAV before 0.97.7 has buffer overflow in the libclamav component

main/http.c in Asterisk Open Source 1.8.x before 1.8.26.1, 11.8.x before 11.8.1, and 12.1.x before 12.1.1, and Certified Asterisk 1.8.x before 1.8.15-cert5 and 11.6 before 11.6-cert2, allows remote attackers to cause a denial of service (stack consumption) and possibly execute arbitrary code via an HTTP request with a large number of Cookie headers.

Multiple off-by-one errors in the (1) MakeBigReq and (2) SetReqLen macros in include/X11/Xlibint.h in X11R6.x and libX11 before 1.6.0 allow remote attackers to have unspecified impact via a crafted request, which triggers a buffer overflow.

Heap-based buffer overflow in the ALGnew function in block_templace.c in Python Cryptography Toolkit (aka pycrypto) allows remote attackers to execute arbitrary code as demonstrated by a crafted iv parameter to cryptmsg.py.

The JavaScript implementation in Mozilla Firefox before 26.0, Firefox ESR 24.x before 24.2, Thunderbird before 24.2, and SeaMonkey before 2.23 does not properly enforce certain typeset restrictions on the generation of GetElementIC typed array stubs, which has unspecified impact and remote attack vectors.

OpenConnect 8.09 has a buffer overflow, causing a denial of service (application crash) or possibly unspecified other impact, via crafted certificate data to get_cert_name in gnutls.c.

Apache HTTP server 2.4.32 to 2.4.44 mod_proxy_uwsgi info disclosure and possible RCE

OpenDMARC through 1.3.2 and 1.4.x through 1.4.0-Beta1 has improper null termination in the function opendmarc_xml_parse that can result in a one-byte heap overflow in opendmarc_xml when parsing a specially crafted DMARC aggregate report. This can cause remote memory corruption when a '\0' byte overwrites the heap metadata of the next chunk and its PREV_INUSE flag.

Use-after-free vulnerability in the nsEventListenerManager::HandleEventSubType function in Mozilla Firefox before 26.0, Firefox ESR 24.x before 24.2, Thunderbird before 24.2, and SeaMonkey before 2.23 allows remote attackers to execute arbitrary code or cause a denial of service (heap memory corruption) via vectors related to mListeners event listeners.

An eval() vulnerability exists in Python Software Foundation Djblets 0.7.21 and Beanbag Review Board before 1.7.15 when parsing JSON requests.

In Twisted Web through 19.10.0, there was an HTTP request splitting vulnerability. When presented with two content-length headers, it ignored the first header. When the second content-length value was set to zero, the request body was interpreted as a pipelined request.

LibreOffice is typically bundled with LibreLogo, a programmable turtle vector graphics script, which can execute arbitrary python commands contained with the document it is launched from. Protection was added, to address CVE-2019-9848, to block calling LibreLogo from document event script handers, e.g. mouse over. However LibreOffice also has a separate feature where documents can specify that pre-installed scripts can be executed on various global script events such as document-open, etc. In the fixed versions, global script event handlers are validated equivalently to document script event handlers. This issue affects: Document Foundation LibreOffice versions prior to 6.2.6.

LibreOffice has a feature where documents can specify that pre-installed scripts can be executed on various document events such as mouse-over, etc. LibreOffice is typically also bundled with LibreLogo, a programmable turtle vector graphics script, which can be manipulated into executing arbitrary python commands. By using the document event feature to trigger LibreLogo to execute python contained within a document a malicious document could be constructed which would execute arbitrary python commands silently without warning. In the fixed versions, LibreLogo cannot be called from a document event handler. This issue affects: Document Foundation LibreOffice versions prior to 6.2.5.

In PuTTY versions before 0.71 on Unix, a remotely triggerable buffer overflow exists in any kind of server-to-client forwarding.

Insufficient sanitization of environment variables passed to rsync can bypass the restrictions imposed by rssh, a restricted shell that should restrict users to perform only rsync operations, resulting in the execution of arbitrary shell commands.

A buffer overflow in DOSBox 0.74-2 allows attackers to execute arbitrary code.