In Puma (RubyGem) before 4.3.5 and 3.12.6, a client could smuggle a request through a proxy, causing the proxy to send a response back to another unknown client. If the proxy uses persistent connections and the client adds another request in via HTTP pipelining, the proxy may mistake it as the first request's body. Puma, however, would see it as two requests, and when processing the second request, send back a response that the proxy does not expect. If the proxy has reused the persistent connection to Puma to send another request for a different client, the second response from the first client will be sent to the second client. This is a similar but different vulnerability from CVE-2020-11076. The problem has been fixed in Puma 3.12.6 and Puma 4.3.5.

In Puma (RubyGem) before 4.3.4 and 3.12.5, an attacker could smuggle an HTTP response, by using an invalid transfer-encoding header. The problem has been fixed in Puma 3.12.5 and Puma 4.3.4.

An issue was discovered in RubyGems 2.6 and later through 3.0.2. Since Gem::CommandManager#run calls alert_error without escaping, escape sequence injection is possible. (There are many ways to cause an error.)

An issue was discovered in RubyGems 2.6 and later through 3.0.2. The gem owner command outputs the contents of the API response directly to stdout. Therefore, if the response is crafted, escape sequence injection may occur.

An integer overflow exists in HAProxy 2.0 through 2.5 in htx_add_header that can be exploited to perform an HTTP request smuggling attack, allowing an attacker to bypass all configured http-request HAProxy ACLs and possibly other ACLs.

An issue was discovered in RubyGems 2.6 and later through 3.0.2. Gem::GemcutterUtilities#with_response may output the API response to stdout as it is. Therefore, if the API side modifies the response, escape sequence injection may occur.

An issue was discovered in RubyGems 2.6 and later through 3.0.2. Since Gem::UserInteraction#verbose calls say without escaping, escape sequence injection is possible.

python-gnupg 0.4.3 allows context-dependent attackers to trick gnupg to decrypt other ciphertext than intended. To perform the attack, the passphrase to gnupg must be controlled by the adversary and the ciphertext should be trusted. Related to a "CWE-20: Improper Input Validation" issue affecting the affect functionality component.

Moodle before 2.2.2 has an external enrolment plugin context check issue where capability checks are not thorough

The pip package before 19.2 for Python allows Directory Traversal when a URL is given in an install command, because a Content-Disposition header can have ../ in a filename, as demonstrated by overwriting the /root/.ssh/authorized_keys file. This occurs in _download_http_url in _internal/download.py.

An issue was discovered in Symfony 2.8.0 through 2.8.50, 3.4.0 through 3.4.34, 4.2.0 through 4.2.11, and 4.3.0 through 4.3.7. If an application passes unvalidated user input as the file for which MIME type validation should occur, then arbitrary arguments are passed to the underlying file command. This is related to symfony/http-foundation (and symfony/mime in 4.3.x).

Matrix Synapse before 0.34.0.1, when the macaroon_secret_key authentication parameter is not set, uses a predictable value to derive a secret key and other secrets which could allow remote attackers to impersonate users.

In MediaWiki before 1.31.15, 1.32.x through 1.35.x before 1.35.3, and 1.36.x before 1.36.1, bots have certain unintended API access. When a bot account has a "sitewide block" applied, it is able to still "purge" pages through the MediaWiki Action API (which a "sitewide block" should have prevented).

An issue was discovered in Squid 3.x and 4.x through 4.8. It allows attackers to smuggle HTTP requests through frontend software to a Squid instance that splits the HTTP Request pipeline differently. The resulting Response messages corrupt caches (between a client and Squid) with attacker-controlled content at arbitrary URLs. Effects are isolated to software between the attacker client and Squid. There are no effects on Squid itself, nor on any upstream servers. The issue is related to a request header containing whitespace between a header name and a colon.

PuTTY before 0.73 mishandles the "bracketed paste mode" protection mechanism, which may allow a session to be affected by malicious clipboard content.

Waitress through version 1.3.1 would parse the Transfer-Encoding header and only look for a single string value, if that value was not chunked it would fall through and use the Content-Length header instead. According to the HTTP standard Transfer-Encoding should be a comma separated list, with the inner-most encoding first, followed by any further transfer codings, ending with chunked. Requests sent with: "Transfer-Encoding: gzip, chunked" would incorrectly get ignored, and the request would use a Content-Length header instead to determine the body size of the HTTP message. This could allow for Waitress to treat a single request as multiple requests in the case of HTTP pipelining. This issue is fixed in Waitress 1.4.0.

runc through 1.0.0-rc8, as used in Docker through 19.03.2-ce and other products, allows AppArmor restriction bypass because libcontainer/rootfs_linux.go incorrectly checks mount targets, and thus a malicious Docker image can mount over a /proc directory.

Dino before 2019-09-10 does not properly check the source of an MAM message in module/xep/0313_message_archive_management.vala.

crypto/x509/x509_vfy.c in OpenSSL 1.0.x before 1.0.0e does not initialize certain structure members, which makes it easier for remote attackers to bypass CRL validation by using a nextUpdate value corresponding to a time in the past.

MediaWiki before 1.35.1 blocks legitimate attempts to hide log entries in some situations. If one sets MediaWiki:Mainpage to Special:MyLanguage/Main Page, visits a log entry on Special:Log, and toggles the "Change visibility of selected log entries" checkbox (or a tags checkbox) next to it, there is a redirection to the main page's action=historysubmit (instead of the desired behavior in which a revision-deletion form appears).

cpio, as used in build 2007.05.10, 2010.07.28, and possibly other versions, allows remote attackers to overwrite arbitrary files via a symlink within an RPM package archive.

Vulnerability in the Java SE product of Oracle Java SE (component: JSSE). Supported versions that are affected are Java SE: 11.0.6 and 14. Easily exploitable vulnerability allows unauthenticated attacker with network access via HTTPS to compromise Java SE. Successful attacks of this vulnerability can result in unauthorized creation, deletion or modification access to critical data or all Java SE accessible data. Note: This vulnerability can only be exploited by supplying data to APIs in the specified Component without using Untrusted Java Web Start applications or Untrusted Java applets, such as through a web service. CVSS 3.0 Base Score 7.5 (Integrity impacts). CVSS Vector: (CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N).

The build package before 20171128 did not check directory names during extraction of build results that allowed untrusted builds to write outside of the target system,allowing escape out of buildroots.

An issue was discovered in the FileImporter extension for MediaWiki before 1.34.4. An attacker can import a file even when the target page is protected against "page creation" and the attacker should not be able to create it. This occurs because of a mishandled distinction between an upload restriction and a create restriction. An attacker cannot leverage this to overwrite anything, but can leverage this to force a wiki to have a page with a disallowed title.

An issue was discovered in Ruby through 2.5.8, 2.6.x through 2.6.6, and 2.7.x through 2.7.1. WEBrick, a simple HTTP server bundled with Ruby, had not checked the transfer-encoding header value rigorously. An attacker may potentially exploit this issue to bypass a reverse proxy (which also has a poor header check), which may lead to an HTTP Request Smuggling attack.

If the upload course tool in Moodle was used to delete an enrollment method which did not exist or was not already enabled, the tool would erroneously enable that enrollment method. This could lead to unintended users gaining access to the course. Versions affected: 3.9 to 3.9.2, 3.8 to 3.8.5, 3.7 to 3.7.8, 3.5 to 3.5.14 and earlier unsupported versions. This is fixed in moodle 3.9.3, 3.8.6, 3.7.9, 3.5.15, and 3.10.

In moodle, insufficient capability checks could lead to users with the ability to course restore adding additional capabilities to roles within that course. Versions affected: 3.9 to 3.9.2, 3.8 to 3.8.5, 3.7 to 3.7.8, 3.5 to 3.5.14 and earlier unsupported versions. This is fixed in moodle 3.9.3, 3.8.6, 3.7.9, 3.5.15, and 3.10.

A flaw was found in FasterXML Jackson Databind, where it did not have entity expansion secured properly. This flaw allows vulnerability to XML external entity (XXE) attacks. The highest threat from this vulnerability is data integrity.

FedMsg 0.18.1 and older is vulnerable to a message validation flaw resulting in message validation not being enabled if configured to be on.

HKDF in cryptography before 1.5.2 returns an empty byte-string if used with a length less than algorithm.digest_size.

The mkdir procedure of GNU Guile temporarily changed the process' umask to zero. During that time window, in a multithreaded application, other threads could end up creating files with insecure permissions. For example, mkdir without the optional mode argument would create directories as 0777. This is fixed in Guile 2.0.13. Prior versions are affected.

elog 3.1.1 allows remote attackers to post data as any username in the logbook.

The route manager in FlightGear before 2016.4.4 allows remote attackers to write to arbitrary files via a crafted Nasal script.

OpenSSL 0.9.4 and OpenSSH for FreeBSD do not properly check for the existence of the /dev/random or /dev/urandom devices, which are absent on FreeBSD Alpha systems, which causes them to produce weak keys which may be more easily broken.

curl and libcurl before 7.50.1 do not check the client certificate when choosing the TLS connection to reuse, which might allow remote attackers to hijack the authentication of the connection by leveraging a previously created connection with a different client certificate.

The socket_create function in common/socket.c in libimobiledevice and libusbmuxd allows remote attackers to bypass intended access restrictions and communicate with services on iOS devices by connecting to an IPv4 TCP socket.

ecdsautils is a tiny collection of programs used for ECDSA (keygen, sign, verify). `ecdsa_verify_[prepare_]legacy()` does not check whether the signature values `r` and `s` are non-zero. A signature consisting only of zeroes is always considered valid, making it trivial to forge signatures. Requiring multiple signatures from different public keys does not mitigate the issue: `ecdsa_verify_list_legacy()` will accept an arbitrary number of such forged signatures. Both the `ecdsautil verify` CLI command and the libecdsautil library are affected. The issue has been fixed in ecdsautils 0.4.1. All older versions of ecdsautils (including versions before the split into a library and a CLI utility) are vulnerable.

Botan 1.11.x before 1.11.29 does not enforce TLS policy for (1) signature algorithms and (2) ECC curves, which allows remote attackers to conduct downgrade attacks via unspecified vectors.

Waitress through version 1.3.1 implemented a "MAY" part of the RFC7230 which states: "Although the line terminator for the start-line and header fields is the sequence CRLF, a recipient MAY recognize a single LF as a line terminator and ignore any preceding CR." Unfortunately if a front-end server does not parse header fields with an LF the same way as it does those with a CRLF it can lead to the front-end and the back-end server parsing the same HTTP message in two different ways. This can lead to a potential for HTTP request smuggling/splitting whereby Waitress may see two requests while the front-end server only sees a single HTTP message. This issue is fixed in Waitress 1.4.0.

Apache HTTP Server versions 2.4.6 to 2.4.46 mod_proxy_wstunnel configured on an URL that is not necessarily Upgraded by the origin server was tunneling the whole connection regardless, thus allowing for subsequent requests on the same connection to pass through with no HTTP validation, authentication or authorization possibly configured.

Go before 1.12.10 and 1.13.x before 1.13.1 allow HTTP Request Smuggling.

Dino before 2019-09-10 does not check roster push authorization in module/roster/module.vala.

Dino before 2019-09-10 does not properly check the source of a carbons message in module/xep/0280_message_carbons.vala.

A vulnerability was discovered in DNS resolver of knot resolver before version 4.1.0 which allows remote attackers to downgrade DNSSEC-secure domains to DNSSEC-insecure state, opening possibility of domain hijack using attacks against insecure DNS protocol.

Puma is a simple, fast, multi-threaded, parallel HTTP 1.1 server for Ruby/Rack applications. When using Puma behind a proxy that does not properly validate that the incoming HTTP request matches the RFC7230 standard, Puma and the frontend proxy may disagree on where a request starts and ends. This would allow requests to be smuggled via the front-end proxy to Puma. The vulnerability has been fixed in 5.6.4 and 4.3.12. Users are advised to upgrade as soon as possible. Workaround: when deploying a proxy in front of Puma, turning on any and all functionality to make sure that the request matches the RFC7230 standard.

Moment.js is a JavaScript date library for parsing, validating, manipulating, and formatting dates. A path traversal vulnerability impacts npm (server) users of Moment.js between versions 1.0.1 and 2.29.1, especially if a user-provided locale string is directly used to switch moment locale. This problem is patched in 2.29.2, and the patch can be applied to all affected versions. As a workaround, sanitize the user-provided locale name before passing it to Moment.js.

Pulp before 2.8.5 uses bash's $RANDOM in an unsafe way to generate passwords.

sf-pcapng.c in libpcap before 1.9.1 does not properly validate the PHB header length before allocating memory.