The freetype project reports:

A heap buffer overflow has been found in the handling of embedded PNG bitmaps, introduced in FreeType version 2.6.

Oracle reports:

This Critical Patch Update contains 48 new security patches for Oracle MySQL.

The highest CVSS v3.1 Base Score of vulnerabilities affecting Oracle MySQL is 8.

Chrome Releases reports:

This release includes 5 security fixes:

• [1125337] High CVE-2020-16000: Inappropriate implementation in Blink. Reported by amaebi_jp on 2020-09-06
• [1135018] High CVE-2020-16001: Use after free in media. Reported by Khalil Zhani on 2020-10-05
• [1137630] High CVE-2020-16002: Use after free in PDFium. Reported by Weipeng Jiang (@Krace) from Codesafe Team of Legendsec at Qi'anxin Group on 2020-10-13
• [1139963] High CVE-2020-15999: Heap buffer overflow in Freetype. Reported by Sergei Glazunov of Google Project Zero on 2020-10-19
• [1134960] Medium CVE-2020-16003: Use after free in printing. Reported by Khalil Zhani on 2020-10-04

The MariaDB project reports:

Details of this vulnerability have not yet been disclosed

Matrix developers reports:

The fallback authentication endpoint served via Synapse were vulnerable to cross-site scripting (XSS) attacks. The impact depends on the configuration of the domain that Synapse is deployed on, but may allow access to cookies and other browser data, CSRF vulnerabilities, and access to other resources served on the same domain or parent domains.

Drupal Security Team reports:

The Drupal AJAX API does not disable JSONP by default, which can lead to cross-site scripting.

PowerDNS Team reports:

CVE-2020-25829: An issue has been found in PowerDNS Recursor where a remote attacker can cause the cached records for a given name to be updated to the ?Bogus? DNSSEC validation state, instead of their actual DNSSEC ?Secure? state, via a DNS ANY query. This results in a denial of service for installations that always validate (dnssec=validate) and for clients requesting validation when on-demand validation is enabled (dnssec=process).

Adobe reports:

• This update resolves a NULL pointer dereference vulnerability that could lead to arbitrary code execution (CVE-2020-9746).

Ruby on Rails blog:

Rails version 6.0.3.4 has been released! This version is a security release and addresses one possible XSS attack vector in Actionable Exceptions.

Chrome releases reports:

This release contains 35 security fixes, including:

• [1127322] Critical CVE-2020-15967: Use after free in payments. Reported by Man Yue Mo of GitHub Security Lab on 2020-09-11
• [1126424] High CVE-2020-15968: Use after free in Blink. Reported by Anonymous on 2020-09-09
• [1124659] High CVE-2020-15969: Use after free in WebRTC. Reported by Anonymous on 2020-09-03
• [1108299] High CVE-2020-15970: Use after free in NFC. Reported by Man Yue Mo of GitHub Security Lab on 2020-07-22
• [1114062] High CVE-2020-15971: Use after free in printing. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2020-08-07
• [1115901] High CVE-2020-15972: Use after free in audio. Reported by Anonymous on 2020-08-13
• [1133671] High CVE-2020-15990: Use after free in autofill. Reported by Rong Jian and Guang Gong of Alpha Lab, Qihoo 360 on 2020-09-30
• [1133688] High CVE-2020-15991: Use after free in password manager. Reported by Rong Jian and Guang Gong of Alpha Lab, Qihoo 360 on 2020-09-30
• [1106890] Medium CVE-2020-15973: Insufficient policy enforcement in extensions. Reported by David Erceg on 2020-07-17
• [1104103] Medium CVE-2020-15974: Integer overflow in Blink. Reported by Juno Im (junorouse) of Theori on 2020-07-10
• [1110800] Medium CVE-2020-15975: Integer overflow in SwiftShader. Reported by Anonymous on 2020-07-29
• [1123522] Medium CVE-2020-15976: Use after free in WebXR. Reported by YoungJoo Lee (@ashuu_lee) of Raon Whitehat on 2020-08-31
• [1083278] Medium CVE-2020-6557: Inappropriate implementation in networking. Reported by Matthias Gierlings and Marcus Brinkmann (NDS Ruhr-University Bochum) on 2020-05-15
• [1097724] Medium CVE-2020-15977: Insufficient data validation in dialogs. Reported by Narendra Bhati (@imnarendrabhati) on 2020-06-22
• [1116280] Medium CVE-2020-15978: Insufficient data validation in navigation. Reported by Luan Herrera (@lbherrera_) on 2020-08-14
• [1127319] Medium CVE-2020-15979: Inappropriate implementation in V8. Reported by Avihay Cohen (@SeraphicAlgorithms) on 2020-09-11
• [1092453] Medium CVE-2020-15980: Insufficient policy enforcement in Intents. Reported by Yongke Wang (@Rudykewang) and Aryb1n (@aryb1n) of Tencent Security Xuanwu Lab on 2020-06-08
• [1123023] Medium CVE-2020-15981: Out of bounds read in audio. Reported by Christoph Guttandin on 2020-08-28
• [1039882] Medium CVE-2020-15982: Side-channel information leakage in cache. Reported by Luan Herrera (@lbherrera_) on 2020-01-07
• [1076786] Medium CVE-2020-15983: Insufficient data validation in webUI. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2020-04-30
• [1080395] Medium CVE-2020-15984: Insufficient policy enforcement in Omnibox. Reported by Rayyan Bijoora on 2020-05-07
• [1099276] Medium CVE-2020-15985: Inappropriate implementation in Blink. Reported by Abdulrahman Alqabandi, Microsoft Browser Vulnerability Research on 2020-06-25
• [1100247] Medium CVE-2020-15986: Integer overflow in media. Reported by Mark Brand of Google Project Zero on 2020-06-29
• [1127774] Medium CVE-2020-15987: Use after free in WebRTC. Reported by Philipp Hancke on 2020-09-14
• [1110195] Medium CVE-2020-15992: Insufficient policy enforcement in networking. Reported by Alison Huffman, Microsoft Browser Vulnerability Research on 2020-07-28
• [1092518] Low CVE-2020-15988: Insufficient policy enforcement in downloads. Reported by Samuel Attard on 2020-06-08
• [1108351] Low CVE-2020-15989: Uninitialized Use in PDFium. Reported by Gareth Evans (Microsoft) on 2020-07-22

Jon Siwek of Corelight reports:

This release fixes the following security issue:

• A memory leak in multipart MIME code has potential for remote exploitation and cause for Denial of Service via resource exhaustion.

Payara Releases reports:

The following is a list of tracked Common Vulnerabilities and Exposures that have been reported and analyzed, which can or have impacted Payara Server across releases:

• CVE-2018-14721 FasterXML jackson-databind 2.x before 2.9.7 might allow remote attackers to conduct server-side request forgery (SSRF) attacks
• CVE-2018-14720 FasterXML jackson-databind 2.x before 2.9.7 might allow remote attackers to conduct external XML entity (XXE) attacks
• CVE-2018-14719 FasterXML jackson-databind 2.x before 2.9.7 might allow remote attackers to execute arbitrary code
• CVE-2018-14718 FasterXML jackson-databind 2.x before 2.9.7 might allow remote attackers to execute arbitrary code
• CVE-2018-14371 Eclipse Mojarra before 2.3.7 is affected by Directory Traversal via the loc parameter

Payara Releases reports:

The following is a list of tracked Common Vulnerabilities and Exposures that have been reported and analyzed, which can or have impacted Payara Server across releases:

• CVE-2020-6950 Eclipse Mojarra vulnerable to path trasversal flaw via either loc/con parameters

Payara Releases reports:

The following is a list of tracked Common Vulnerabilities and Exposures that have been reported and analyzed, which can or have impacted Payara Server across releases:

• CVE-2019-12086 A Polymorphic Typing issue was discovered in FasterXML jackson-databind 2.x before 2.9.9

Release notes:

Lots of fixes exposed by fuzzers like AFL, ClusterFuzz, OSSFuzz and others:

CVE-2016-6328: fixed integer overflow when parsing maker notes

CVE-2017-7544: fixed buffer overread

CVE-2018-20030: Fix for recursion DoS

CVE-2019-9278: replaced integer overflow checks the compiler could optimize away by safer constructs

CVE-2020-0093: read overflow

CVE-2020-12767: fixed division by zero

CVE-2020-13112: Various buffer overread fixes due to integer overflows in maker notes

CVE-2020-13113: Potential use of uninitialized memory

CVE-2020-13114: Time consumption DoS when parsing canon array markers

Albert Astals Cid reports:

KDE Project Security Advisory

Title	KDE Connect: packet manipulation can be exploited in a Denial of Service attack
Risk Rating	Important
CVE	CVE-2020-26164
Versions	kdeconnect <= 20.08.1
Author	Albert Vaca Cintora
Date	2 October 2020

Overview

An attacker on your local network could send maliciously crafted packets to other hosts running kdeconnect on the network, causing them to use large amounts of CPU, memory or network connections, which could be used in a Denial of Service attack within the network.

Impact

Computers that run kdeconnect are susceptible to DoS attacks from the local network.

Workaround

We advise you to stop KDE Connect when on untrusted networks like those on airports or conferences.

Since kdeconnect is dbus activated it is relatively hard to make sure it stays stopped so the brute force approach is to uninstall the kdeconnect package from your system and then run

	      kquitapp5 kdeconnectd
	  

Just install the package again once you're back in a trusted network.

Solution

KDE Connect 20.08.2 patches several code paths that could result in a DoS.

You can apply these patches on top of 20.08.1:

• https://invent.kde.org/network/kdeconnect-kde/-/commit/f183b5447bad47655c21af87214579f03bf3a163
• https://invent.kde.org/network/kdeconnect-kde/-/commit/b279c52101d3f7cc30a26086d58de0b5f1c547fa
• https://invent.kde.org/network/kdeconnect-kde/-/commit/d35b88c1b25fe13715f9170f18674d476ca9acdc
• https://invent.kde.org/network/kdeconnect-kde/-/commit/b496e66899e5bc9547b6537a7f44ab44dd0aaf38
• https://invent.kde.org/network/kdeconnect-kde/-/commit/5310eae85dbdf92fba30375238a2481f2e34943e
• https://invent.kde.org/network/kdeconnect-kde/-/commit/721ba9faafb79aac73973410ee1dd3624ded97a5
• https://invent.kde.org/network/kdeconnect-kde/-/commit/ae58b9dec49c809b85b5404cee17946116f8a706
• https://invent.kde.org/network/kdeconnect-kde/-/commit/66c768aa9e7fba30b119c8b801efd49ed1270b0a
• https://invent.kde.org/network/kdeconnect-kde/-/commit/85b691e40f525e22ca5cc4ebe79c361d71d7dc05
• https://invent.kde.org/network/kdeconnect-kde/-/commit/48180b46552d40729a36b7431e97bbe2b5379306

Credits

Thanks Matthias Gerstner and the openSUSE security team for reporting the issue.

Thanks to Aleix Pol, Nicolas Fella and Albert Vaca Cintora for the patches.

CVE mitre reports:

Portable UPnP SDK (aka libupnp) 1.12.1 and earlier allows remote attackers to cause a denial of service (crash) via a crafted SSDP message due to a NULL pointer dereference in the functions FindServiceControlURLPath and FindServiceEventURLPath in genlib/service_table/service_table.c.

Gitlab reports:

Potential Denial Of Service Via Update Release Links API

Insecure Storage of Session Key In Redis

Improper Access Expiration Date Validation

Cross-Site Scripting in Multiple Pages

Unauthorized Users Can View Custom Project Template

Cross-Site Scripting in SVG Image Preview

Incomplete Handling in Account Deletion

Insufficient Rate Limiting at Re-Sending Confirmation Email

Improper Type Check in GraphQL

To-dos Are Not Redacted When Membership Changes

Guest users can modify confidentiality attribute

Command injection on runner host

Insecure Runner Configuration in Kubernetes Environments

Apache reports:

Apache Ant 1.1 to 1.9.14 and 1.10.0 to 1.10.7 uses the default temporary directory identified by the Java system property java.io.tmpdir for several tasks and may thus leak sensitive information. The fixcrlf and replaceregexp tasks also copy files from the temporary directory back into the build tree allowing an attacker to inject modified source files into the build process.

Gnome Pango 1.42 and later is affected by: Buffer Overflow. The impact is: The heap based buffer overflow can be used to get code execution. The component is: function name: pango_log2vis_get_embedding_levels, assignment of nchars and the loop condition. The attack vector is: Bug can be used when application pass invalid utf-8 strings to functions like pango_itemize.

PowerDNS Team reports

CVE-2020-17482: An issue has been found in PowerDNS Authoritative Server before 4.3.1 where an authorized user with the ability to insert crafted records into a zone might be able to leak the content of uninitialized memory. Such a user could be a customer inserting data via a control panel, or somebody with access to the REST API. Crafted records cannot be inserted via AXFR.

Chrome Releases reports:

This release fixes 10 security issues, including:

• [1100136] High CVE-2020-15960: Out of bounds read in storage. Reported by Anonymous on 2020-06-28
• [1114636] High CVE-2020-15961: Insufficient policy enforcement in extensions. Reported by David Erceg on 2020-08-10
• [1121836] High CVE-2020-15962: Insufficient policy enforcement in serial. Reported by Leecraso and Guang Gong of 360 Alpha Lab working with 360 BugCloud on 2020-08-26
• [1113558] High CVE-2020-15963: Insufficient policy enforcement in extensions. Reported by David Erceg on 2020-08-06
• [1126249] High CVE-2020-15965: Out of bounds write in V8. Reported by Lucas Pinheiro, Microsoft Browser Vulnerability Research on 2020-09-08
• [1113565] Medium CVE-2020-15966: Insufficient policy enforcement in extensions. Reported by David Erceg on 2020-08-06
• [1121414] Low CVE-2020-15964: Insufficient data validation in media. Reported by Woojin Oh(@pwn_expoit) of STEALIEN on 2020-08-25

CVE mitre reports:

CVE-2019-20388

xmlSchemaPreRun in xmlschemas.c in libxml2 2.9.10 allows an xmlSchemaValidateStream memory leak.

CVE-2020-7595

xmlStringLenDecodeEntities in parser.c in libxml2 2.9.10 has an infinite loop in a certain end-of-file situation.

CVE-2020-24977

GNOME project libxml2 v2.9.10 and earlier have a global buffer over-read vulnerability in xmlEncodeEntitiesInternal

Problem Description:

Affected Synapse versions assume that all events have an "origin" field set. If an event without the "origin" field is sent into a federated room, servers not already joined to the room will be unable to do so due to failing to fetch the malformed event.

Impact:

An attacker could cause a denial of service by deliberately sending a malformed event into a room, thus preventing new servers (and thus their users) from joining the room.

The Samba Team reports:

An unauthenticated attacker on the network can gain administrator access by exploiting a netlogon protocol flaw.

Python reports:

bpo-39603: Prevent http header injection by rejecting control characters in http.client.putrequest(?).

bpo-29778: Ensure python3.dll is loaded from correct locations when Python is embedded (CVE-2020-15523).

bpo-41004: CVE-2020-14422: The __hash__() methods of ipaddress.IPv4Interface and ipaddress.IPv6Interface incorrectly generated constant hash values of 32 and 128 respectively. This resulted in always causing hash collisions. The fix uses hash() to generate hash values for the tuple of (address, mask length, network address).

bpo-39073: Disallow CR or LF in email.headerregistry.Address arguments to guard against header injection attacks.

bpo-38576: Disallow control characters in hostnames in http.client, addressing CVE-2019-18348. Such potentially malicious header injection URLs now cause a InvalidURL to be raised.

bpo-39503: CVE-2020-8492: The AbstractBasicAuthHandler class of the urllib.request module uses an inefficient regular expression which can be exploited by an attacker to cause a denial of service. Fix the regex to prevent the catastrophic backtracking. Vulnerability reported by Ben Caller and Matt Schwager.

bpo-38945: Newline characters have been escaped when performing uu encoding to prevent them from overflowing into to content section of the encoded file. This prevents malicious or accidental modification of data during the decoding process.

bpo-38804: Fixes a ReDoS vulnerability in http.cookiejar. Patch by Ben Caller.

bpo-39017: Avoid infinite loop when reading specially crafted TAR files using the tarfile module (CVE-2019-20907).

bpo-41183: Use 3072 RSA keys and SHA-256 signature for test certs and keys.

bpo-39503: AbstractBasicAuthHandler of urllib.request now parses all WWW-Authenticate HTTP headers and accepts multiple challenges per header: use the realm of the first Basic challenge.

tt-rss project reports:

The cached_url feature mishandles JavaScript inside an SVG document.

imgproxy in plugins/af_proxy_http/init.php mishandles $_REQUEST["url"] in an error message.

It does not validate all URLs before requesting them.

Allows remote attackers to execute arbitrary PHP code via a crafted plural forms header.

The Nextcloud project reports:

NC-SA-2020-026 (low): Password of share by mail is not hashed when given on the create share call

A logic error in Nextcloud Server 19.0.0 caused a plaintext storage of the share password when it was given on the initial create API call.

Problem Description:

AMD and Intel CPUs support hardware virtualization using specialized data structures that control various aspects of guest operation. These are the Virtual Machine Control Structure (VMCS) on Intel CPUs, and the Virtual Machine Control Block (VMCB) on AMD CPUs. Insufficient access controls allow root users, including those running in a jail, to change these data structures.

Impact:

An attacker with host root access (including to a jailed bhyve instance) can use this vulnerability to achieve kernel code execution.

Node.js reports:

Updates are now available for v10,x, v12.x and v14.x Node.js release lines for the following issues.

HTTP Request Smuggling due to CR-to-Hyphen conversion (High) (CVE-2020-8201)

Affected Node.js versions converted carriage returns in HTTP request headers to a hyphen before parsing. This can lead to HTTP Request Smuggling as it is a non-standard interpretation of the header.

Impacts:

• All versions of the 14.x and 12.x releases line

Denial of Service by resource exhaustion CWE-400 due to unfinished HTTP/1.1 requests (Critical) (CVE-2020-8251)

Node.js is vulnerable to HTTP denial of service (DOS) attacks based on delayed requests submission which can make the server unable to accept new connections. The fix a new http.Server option called requestTimeout with a default value of 0 which means it is disabled by default. This should be set when Node.js is used as an edge server, for more details refer to the documentation.

Impacts:

• All versions of the 14.x release line

fs.realpath.native on may cause buffer overflow (Medium) (CVE-2020-8252)

libuv's realpath implementation incorrectly determined the buffer size which can result in a buffer overflow if the resolved path is longer than 256 bytes.

Impacts:

• All versions of the 10.x release line
• All versions of the 12.x release line
• All versions of the 14.x release line before 14.9.0

Problem Description:

A ftpd(8) bug in the implementation of the file system sandbox, combined with capabilities available to an authenticated FTP user, can be used to escape the file system restriction configured in ftpchroot(5). Moreover, the bug allows a malicious client to gain root privileges.

Impact:

A malicious FTP user can gain privileged access to an affected system.

Problem Description:

A programming error in the ure(4) device driver caused some Realtek USB Ethernet interfaces to incorrectly report packets with more than 2048 bytes in a single USB transfer as having a length of only 2048 bytes.

An adversary can exploit this to cause the driver to misinterpret part of the payload of a large packet as a separate packet, and thereby inject packets across security boundaries such as VLANs.

Impact:

An attacker that can send large frames (larger than 2048 bytes in size) to be received by the host (be it VLAN, or non-VLAN tagged packet), can inject arbitrary packets to be received and processed by the host. This includes spoofing packets from other hosts, or injecting packets to other VLANs than the host is on.

Problem Description:

A number of AMD virtualization instructions operate on host physical addresses, are not subject to nested page table translation, and guest use of these instructions was not trapped.

Impact:

From kernel mode a malicious guest can write to arbitrary host memory (with some constraints), affording the guest full control of the host.

Ruby on Rails blog:

Rails 5.2.4.4 and 6.0.3.3 have been released! These releases contain an important security fix, so please upgrade when you can.

Both releases contain the following fix: [CVE-2020-15169] Potential XSS vulnerability in Action View

Jon Siwek of Corelight reports:

This release fixes the following security issue:

• The AYIYA and GTPv1 parsing/decapsulation logic may leak memory -- These leaks have potential for remote exploitation to cause Denial of Service via resource exhaustion.

Chrome Releases reports:

This release contains 5 security fixes:

• [1116304] High CVE-2020-6573: Use after free in video. Reported by Leecraso and Guang Gong of 360 Alpha Lab working with 360 BugCloud on 2020-08-14
• [1102196] High CVE-2020-6574: Insufficient policy enforcement in installer. Reported by CodeColorist of Ant-Financial LightYear Labs on 2020-07-05
• [1081874] High CVE-2020-6575: Race in Mojo. Reported by Microsoft on 2020-05-12
• [1111737] High CVE-2020-6576: Use after free in offscreen canvas. Reported by Looben Yang on 2020-07-31
• [1122684] High CVE-2020-15959: Insufficient policy enforcement in networking. Reported by Eric Lawrence of Microsoft on 2020-08-27

Chen Nan of Chaitin Security Research Lab reports:

Fix buffer overflow introduced in version 5.8: processing of template %aX in a RADIUS authentication response might lead to unexpected termination of the mpd5 process. Installations not using RADIUS or not using %aX templates in RADIUS attributes were not affected.

Fix buffer overflow in parsing of L2TP control packets introduced in version 4.0 that initially brought in L2TP support: a specially crafted incoming L2TP control packet might lead to unexpected termination of the process. Installations with neither L2TP clients nor L2TP server configured are not affected.

Version 5.9 contains security fix for L2TP clients and servers. Insufficient validation of incoming L2TP control packet specially crafted by unauthenticated user might lead to unexpected termination of the process. The problem affects mpd versions since 4.0 that brought in initial support for L2TP. Installations not using L2TP clients nor L2TP server configuration were not affected.

The GnuTLS project reports:

It was found by oss-fuzz that the server sending a "no_renegotiation" alert in an unexpected timing, followed by an invalid second handshake can cause a TLS 1.3 client to crash via a null-pointer dereference. The crash happens in the application's error handling path, where the gnutls_deinit function is called after detecting a handshake failure.

Manuel Pégourié-Gonnard reports:

When decrypting/authenticating (D)TLS record in a connection using a CBC ciphersuite without the Encrypt-then-Mac extension RFC 7366, Mbed TLS used dummy rounds of the compression function associated with the hash used for HMAC in order to hide the length of the padding to remote attackers, as recommended in the original Lucky Thirteen paper.

A local attacker who is able to observe the state of the cache could monitor the presence of mbedtls_md_process() in the cache in order to determine when the actual computation ends and when the dummy rounds start. This is a reliable target as it's always called at least once, in response to a previous attack. The attacker can then continue with one of many well-documented Lucky 13 variants.

Manuel Pégourié-Gonnard reports:

An attacker with access to precise enough timing and memory access information (typically an untrusted operating system attacking a secure enclave such as SGX or the TrustZone secure world) can recover the private keys used in RSA or static (finite-field) Diffie-Hellman operations.

Django Release notes:

CVE-2020-24583: Incorrect permissions on intermediate-level directories on Python 3.7+

On Python 3.7+, FILE_UPLOAD_DIRECTORY_PERMISSIONS mode was not applied to intermediate-level directories created in the process of uploading files and to intermediate-level collected static directories when using the collectstatic management command.

CVE-2020-24584: Permission escalation in intermediate-level directories of the file system cache on Python 3.7+

On Python 3.7+, the intermediate-level directories of the file system cache had the system's standard umask rather than 0o077 (no group or others permissions).

JasPer NEWS:

- Fix CVE-2018-9154

- Fix CVE-2018-19541

- Fix CVE-2016-9399, CVE-2017-13751

- Fix CVE-2018-19540

- Fix CVE-2018-9055

- Fix CVE-2017-13748

- Fix CVE-2017-5503, CVE-2017-5504, CVE-2017-5505

- Fix CVE-2018-9252

- Fix CVE-2018-19139

- Fix CVE-2018-19543, CVE-2017-9782

- Fix CVE-2018-20570

- Fix CVE-2018-20622

- Fix CVE-2016-9398

- Fix CVE-2017-14132

- Fix CVE-2017-5499

- Fix CVE-2018-18873

- Fix CVE-2017-13750

Importing an OpenPGP key having a preference list for AEAD algorithms will lead to an array overflow and thus often to a crash or other undefined behaviour.

Importing an arbitrary key can often easily be triggered by an attacker and thus triggering this bug. Exploiting the bug aside from crashes is not trivial but likely possible for a dedicated attacker. The major hurdle for an attacker is that only every second byte is under their control with every first byte having a fixed value of 0x04.

Gitlab reports:

Vendor Cross-Account Assume-Role Attack

Stored XSS on the Vulnerability Page

Outdated Job Token Can Be Reused to Access Unauthorized Resources

File Disclosure Via Workhorse File Upload Bypass

Unauthorized Maintainer Can Edit Group Badge

Denial of Service Within Wiki Functionality

Sign-in Vulnerable to Brute-force Attacks

Invalidated Session Allows Account Access With an Old Password

GitLab Omniauth Endpoint Renders User Controlled Messages

Blind SSRF Through Repository Mirroring

Information Disclosure Through Incorrect Group Permission Verifications

No Rate Limit on GitLab Webhook Feature

GitLab Session Revocation Feature Does Not Invalidate All Sessions

OAuth Authorization Scope for an External Application Can Be Changed Without User Consent

Unauthorized Maintainer Can Delete Repository

Improper Verification of Deploy-Key Leads to Access Restricted Repository

Disabled Repository Still Accessible With a Deploy-Token

Duplicated Secret Code Generated by 2 Factor Authentication Mechanism

Lack of Validation Within Project Invitation Flow

Current Sessions Not Invalidated Upon Enabling 2 Factor Authentication

Users Without 2 Factor Authentication Can Be Blocked Accessing GitLab

Lack of Upper Bound Check Leading to Possible Denial of Service

2 Factor Authentication for Groups Was Not Enforced Within API Endpoint

GitLab Runner Denial of Service via CI Jobs

Update jQuery Dependency

Problem Description:

Due to improper mbuf handling in the kernel, a use-after-free bug might be triggered by sending IPv6 Hop-by-Hop options over the loopback interface.

Impact:

Triggering the use-after-free situation may result in unintended kernel behaviour including a kernel panic.

Problem Description:

When parsing option 119 data, dhclient(8) computes the uncompressed domain list length so that it can allocate an appropriately sized buffer to store the uncompressed list. The code to compute the length failed to handle certain malformed input, resulting in a heap overflow when the uncompressed list is copied into in inadequately sized buffer.

Impact:

The heap overflow could in principle be exploited to achieve remote code execution. The affected process runs with reduced privileges in a Capsicum sandbox, limiting the immediate impact of an exploit. However, it is possible the bug could be combined with other vulnerabilities to escape the sandbox.

Problem Description:

Due to improper handling in the kernel, a use-after-free bug can be triggered by sending large user messages from multiple threads on the same socket.

Impact:

Triggering the use-after-free situation may result in unintended kernel behaviour including a kernel panic.

The Go project reports:

When a Handler does not explicitly set the Content-Type header, both CGI implementations default to ?text/html?. If an attacker can make a server generate content under their control (e.g. a JSON containing user data or an uploaded image file) this might be mistakenly returned by the server as ?text/html?. If a victim visits such a page they could get the attacker's code executed in the context of the server origin. If an attacker can make a server generate content under their control (e.g. a JSON containing user data or an uploaded image file) this might be mistakenly returned by the server as ?text/html?. If a victim visits such a page they could get the attacker's code executed in the context of the server origin.

Albert Astals Cid reports:

Overview

A maliciously crafted TAR archive containing symlink entries would install files anywhere in the user's home directory upon extraction.

Proof of concept

For testing, an example of malicious archive can be found at dirsymlink.tar

Impact

Users can unwillingly install files like a modified .bashrc, or a malicious script placed in ~/.config/autostart.

Workaround

Before extracting a downloaded archive using the Ark GUI, users should inspect it to make sure it doesn't contain symlink entries pointing outside the extraction folder.

The 'Extract' context menu from the Dolphin file manager shouldn't be used.

Solution

Ark 20.08.1 skips maliciously crafted symlinks when extracting TAR archives.

Alternatively, 8bf8c5ef07b0ac5e914d752681e470dea403a5bd can be applied to previous releases.

Credits

Thanks to Fabian Vogt for reporting this issue and for fixing it.

grigoritchy at gmail dot com reports:

The phar_parse_zipfile function had use-after-free vulnerability because of mishandling of the actual_alias variable.

Chrome Releases reports:

This update includes 20 security fixes, including:

• [1109120] High CVE-2020-6558: Insufficient policy enforcement in iOS. Reported by Alison Huffman, Microsoft Browser Vulnerability Research on 2020-07-24
• [1116706] High CVE-2020-6559: Use after free in presentation API. Reported by Liu Wei and Wu Zekai of Tencent Security Xuanwu Lab on 2020-08-15
• [1108181] Medium CVE-2020-6560: Insufficient policy enforcement in autofill. Reported by Nadja Ungethuem from www.unnex.de on 2020-07-22
• [932892] Medium CVE-2020-6561: Inappropriate implementation in Content Security Policy. Reported by Rob Wu on 2019-02-16
• [1086845] Medium CVE-2020-6562: Insufficient policy enforcement in Blink. Reported by Masato Kinugawa on 2020-05-27
• [1104628] Medium CVE-2020-6563: Insufficient policy enforcement in intent handling. Reported by Pedro Oliveira on 2020-07-12
• [841622] Medium CVE-2020-6564: Incorrect security UI in permissions. Reported by Khalil Zhani on 2018-05-10
• [1029907] Medium CVE-2020-6565: Incorrect security UI in Omnibox. Reported by Khalil Zhani on 2019-12-02
• [1065264] Medium CVE-2020-6566: Insufficient policy enforcement in media. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2020-03-27
• [937179] Low CVE-2020-6567: Insufficient validation of untrusted input in command line handling. Reported by Joshua Graham of TSS on 2019-03-01
• [1092451] Low CVE-2020-6568: Insufficient policy enforcement in intent handling. Reported by Yongke Wang(@Rudykewang) and Aryb1n(@aryb1n) of Tencent Security Xuanwu Lab on 2020-06-08
• [995732] Low CVE-2020-6569: Integer overflow in WebUSB. Reported by guaixiaomei on 2019-08-20
• [1084699] Low CVE-2020-6570: Side-channel information leakage in WebRTC. Reported by Signal/Tenable on 2020-05-19
• [1085315] Low CVE-2020-6571: Incorrect security UI in Omnibox. Reported by Rayyan Bijoora on 2020-05-21

The X.org project reports:

There is an integer overflow and a double free vulnerability in the way LibX11 handles locales. The integer overflow is a necessary precursor to the double free.

Gitlab reports:

Arbitrary File Read when Moving an Issue

Memory Exhaustion via Excessive Logging of Invite Email Error

Denial of Service Through Project Import Feature

User Controlled Git Configuration Settings Resulting in SSRF

Stored XSS in Issue Reference Number Tooltip

Stored XSS in Issues List via Milestone Title

Improper Access Control After Group Transfer

Bypass Email Verification Required for OAuth Flow

Confusion When Using Hexadecimal Branch Names

Insufficient OAuth Revocation

Improper Access Control for Project Sharing

Stored XSS in Jobs Page

Improper Access Control of Applications Page

SSRF into Shared Runner

Update Kramdown Gem

The X.org project reports:

All theses issuses can lead to local privileges elevation on systems where the X server is running privileged.

The handler for the XkbSetNames request does not validate the request length before accessing its contents.

An integer underflow exists in the handler for the XIChangeHierarchy request.

An integer underflow exist in the handler for the XkbSelectEvents request.

An integer underflow exist in the handler for the CreateRegister request of the X record extension.

Miroslav Lichvar reports:

chrony-3.5.1 [...] fixes a security issue in writing of the pidfile.

When chronyd is configured to save the pidfile in a directory where the chrony user has write permissions (e.g. /var/run/chrony - the default since chrony-3.4), an attacker that compromised the chrony user account could create a symbolic link at the location of the pidfile to make chronyd starting with root privileges follow the symlink and write its process ID to a file for which the chrony user doesn't have write permissions, causing a denial of service, or data loss.

This issue was reported by Matthias Gerstner of SUSE.

Ian Jackson and the adns project reports:

Vulnerable applications: all adns callers. Exploitable by: the local recursive resolver. Likely worst case: Remote code execution.

Vulnerable applications: those that make SOA queries. Exploitable by: upstream DNS data sources. Likely worst case: DoS (crash of the adns-using application)

Vulnerable applications: those that use adns_qf_quoteok_query. Exploitable by: sources of query domain names. Likely worst case: DoS (crash of the adns-using application)

Vulnerable applications: adnshost. Exploitable by: code responsible for framing the input. Likely worst case: DoS (adnshost crashes at EOF).

Andrew Walker reports:

Issue 1:

Users are always granted permissions to cd into a directory. The check for whether execute is present on directories is a de-facto no-op. This cannot be mitigated without upgrading. Even setting an explicit "deny - execute" NFSv4 ACE will be bypassed.

Issue 2:

All ACEs for the owner_group (group@) and regular groups (group:) are granted the current user. This means that POSIX mode 770 is de-facto 777, and the below ACL is also de-facto 777 because the groupmember check for builtin_administrators returns True.

root@TESTBOX[~]# getfacl testfile
# file: testfile
# owner: root
# group: wheel
group:builtin_administrators:rwxpDdaARWcCos:-------:allow
	  

Elastic reports:

A field disclosure flaw was found in Elasticsearch when running a scrolling search with Field Level Security. If a user runs the same query another more privileged user recently ran, the scrolling search can leak fields that should be hidden. This could result in an attacker gaining additional permissions against a restricted index.

Python reports:

bpo-29778: Ensure python3.dll is loaded from correct locations when Python is embedded (CVE-2020-15523).

bpo-41004: CVE-2020-14422: The __hash__() methods of ipaddress.IPv4Interface and ipaddress.IPv6Interface incorrectly generated constant hash values of 32 and 128 respectively. This resulted in always causing hash collisions. The fix uses hash() to generate hash values for the tuple of (address, mask length, network address).

bpo-39603: Prevent http header injection by rejecting control characters in http.client.putrequest(...).

curl security problems:

CVE-2020-8231: wrong connect-only connection

An application that performs multiple requests with libcurl's multi API and sets the CURLOPT_CONNECT_ONLY option, might in rare circumstances experience that when subsequently using the setup connect-only transfer, libcurl will pick and use the wrong connection - and instead pick another one the application has created since then.

CURLOPT_CONNECT_ONLY is the option to tell libcurl to not perform an actual transfer, only connect. When that operation is completed, libcurl remembers which connection it used for that transfer and "easy handle". It remembers the connection using a pointer to the internal connectdata struct in memory.

If more transfers are then done with the same multi handle before the connect-only connection is used, leading to the initial connect-only connection to get closed (for example due to idle time-out) while also new transfers (and connections) are setup, such a new connection might end up getting the exact same memory address as the now closed connect-only connection.

If after those operations, the application then wants to use the original transfer's connect-only setup to for example use curl_easy_send() to send raw data over that connection, libcurl could erroneously find an existing connection still being alive at the address it remembered since before even though this is now a new and different connection.

The application could then accidentally send data over that connection which wasn't at all intended for that recipient, entirely unknowingly.

Icinga development team reports:

CVE-2020-24368

Icinga Icinga Web2 2.0.0 through 2.6.4, 2.7.4 and 2.8.2 has a Directory Traversal vulnerability which allows an attacker to access arbitrary files that are readable by the process running Icinga Web 2. This issue is fixed in Icinga Web 2 in v2.6.4, v2.7.4 and v2.8.2.

Chrome Releases reports:

This release contains one security fix:

• [1115345] High CVE-2020-6556: Heap buffer overflow in SwiftShader. Reported by Alison Huffman, Microsoft Browser Vulnerability Research on 2020-08-12

the TrouSerS project reports reports:

If the tcsd daemon is started with root privileges, it fails to drop the root gid after it is no longer needed.

If the tcsd daemon is started with root privileges, the tss user has read and write access to the /etc/tcsd.conf file.

If the tcsd daemon is started with root privileges, the creation of the system.data file is prone to symlink attacks.

Jenkins Security Advisory:

Description

(Critical) SECURITY-1983 / CVE-2019-17638

Buffer corruption in bundled Jetty

rsync developers reports:

Various zlib fixes, including security fixes for CVE-2016-9843, CVE-2016-9842, CVE-2016-9841, and CVE-2016-9840

py-ecdsa developers report:

Fix CVE-2019-14853 - possible DoS caused by malformed signature decoding.

Fix CVE-2019-14859 - signature malleability caused by insufficient checks of DER encoding

Red Hat bugzilla reports:

A flaw was found in the Red Hat Ceph Storage RadosGW (Ceph Object Gateway). The vulnerability is related to the injection of HTTP headers via a CORS ExposeHeader tag. The newline character in the ExposeHeader tag in the CORS configuration file generates a header injection in the response when the CORS request is made.

Snmptt reports:

Fixed a security issue with EXEC / PREXEC / unknown_trap_exec that could allow malicious shell code to be executed.

Fixed a bug with EXEC / PREXEC / unknown_trap_exec that caused commands to be run as root instead of the user defined in daemon_uid.

Aki Tuomi reports:

Parsing mails with a large number of MIME parts could have resulted in excessive CPU usage or a crash due to running out of stack memory..

Dovecot's NTLM implementation does not correctly check message buffer size, which leads to reading past allocation which can lead to crash

lmtp/submission: Issuing the RCPT command with an address that has the empty quoted string as local-part causes the lmtp service to crash.

Dovecot's RPA mechanism implementation accepts zero-length message, which leads to assert-crash later on.

Cary Phillips reports:

v2.5.3 - Patch release with various bug/security fixes [...]:

• Various sanitizer/fuzz-identified issues related to handling of invalid input

Jenkins Security Advisory:

Description

(High) SECURITY-1955 / CVE-2020-2229

Stored XSS vulnerability in help icons

(High) SECURITY-1957 / CVE-2020-2230

Stored XSS vulnerability in project naming strategy

(High) SECURITY-1960 / CVE-2020-2231

Stored XSS vulnerability in 'Trigger builds remotely'

Puppetlabs reports:

In June 2020, jackson-databind published security updates addressing several CVEs. Previous releases of PuppetDB contain a vulnerable version of jackson.core:jackson-databind. PuppetDB 5.2.18 contains an updated version of jackson-databind that has patched the vulnerabilities.

Chrome Releases reports:

This release contains 15 security fixes, including:

• [1107433] High CVE-2020-6542: Use after free in ANGLE. Reported by Piotr Bania of Cisco Talos on 2020-07-20
• [1104046] High CVE-2020-6543: Use after free in task scheduling. Reported by Looben Yang on 2020-07-10
• [1108497] High CVE-2020-6544: Use after free in media. Reported by Tim Becker of Theori on 2020-07-22
• [1095584] High CVE-2020-6545: Use after free in audio. Reported by Anonymous on 2020-06-16
• [1100280] High CVE-2020-6546: Inappropriate implementation in installer. Reported by Andrew Hess (any1) on 2020-06-29
• [1102153] High CVE-2020-6547: Incorrect security UI in media. Reported by David Albert on 2020-07-05
• [1103827] High CVE-2020-6548: Heap buffer overflow in Skia. Reported by Choongwoo Han, Microsoft Browser Vulnerability Research on 2020-07-09
• [1105426] High CVE-2020-6549: Use after free in media. Reported by Sergei Glazunov of Google Project Zero on 2020-07-14
• [1106682] High CVE-2020-6550: Use after free in IndexedDB. Reported by Sergei Glazunov of Google Project Zero on 2020-07-17
• [1107815] High CVE-2020-6551: Use after free in WebXR. Reported by Sergei Glazunov of Google Project Zero on 2020-07-21
• [1108518] High CVE-2020-6552: Use after free in Blink. Reported by Tim Becker of Theori on 2020-07-22
• [1111307] High CVE-2020-6553: Use after free in offline mode. Reported by Alison Huffman, Microsoft Browser Vulnerability Research on 2020-07-30
• [1094235] Medium CVE-2020-6554: Use after free in extensions. Reported by Anonymous on 2020-06-12
• [1105202] Medium CVE-2020-6555: Out of bounds read in WebGL. Reported by Marcin Towalski of Cisco Talos on 2020-07-13

Bftpd project reports:

Bftpd is vulnerable to out of bounds memory access, file descriptor leak and a potential buffer overflow.

The Apache httpd projec reports:

• mod_http2: Important: Push Diary Crash on Specifically Crafted HTTP/2 Header (CVE-2020-9490)
  
  A specially crafted value for the 'Cache-Digest' header in a HTTP/2 request would result in a crash when the server actually tries to HTTP/2 PUSH a resource afterwards.
• mod_proxy_uwsgi: Moderate: mod_proxy_uwsgi buffer overflow (CVE-2020-11984)
  
  info disclosure and possible RCE
• mod_http2: Moderate: Push Diary Crash on Specifically Crafted HTTP/2 Header (CVE-2020-11993)
  
  When trace/debug was enabled for the HTTP/2 module and on certain traffic edge patterns, logging statements were made on the wrong connection, causing concurrent use of memory pools.

Problem Description:

When handling a 32-bit sendmsg(2) call, the compat32 subsystem copies the control message to be transmitted (if any) into kernel memory, and adjusts alignment of control message headers. The code which performs this work contained a time-of-check to time-of-use (TOCTOU) vulnerability which allows a malicious userspace program to modify control message headers after they were validated by the kernel.

Impact:

The TOCTOU bug can be exploited by an unprivileged malicious userspace program to trigger privilege escalation.

Problem Description:

A missing length validation code common to these three drivers means that a malicious USB device could write beyond the end of an allocated network packet buffer.

Impact:

An attacker with physical access to a USB port and the ability to bring a network interface up may be able to use a specially crafted USB device to gain kernel or user-space code execution.

The Go project reports:

Certain invalid inputs to ReadUvarint or ReadVarint could cause those functions to read an unlimited number of bytes from the ByteReader argument before returning an error. This could lead to processing more input than expected when the caller is reading directly from the network and depends on ReadUvarint and ReadVarint only consuming a small, bounded number of bytes, even from invalid inputs.

sqlite3 update:

Various security issues could be used by an attacker to cause SQLite to crash, resulting in a denial of service, or possibly execute arbitrary code.

• CVE-2020-11655: SQLite through 3.31.1 allows attackers to cause a denial of service (segmentation fault) via a malformed window-function query because the AggInfo object's initialization is mishandled.
• CVE-2020-13434: SQLite through 3.32.0 has an integer overflow in sqlite3_str_vappendf in printf.c.
• CVE-2020-13435: SQLite through 3.32.0 has a segmentation fault in sqlite3ExprCodeTarget in expr.c.
• CVE-2020-13630: ext/fts3/fts3.c in SQLite before 3.32.0 has a use-after-free in fts3EvalNextRow, related to the snippet feature.
• CVE-2020-13631: SQLite before 3.32.0 allows a virtual table to be renamed to the name of one of its shadow tables, related to alter.c and build.c.
• CVE-2020-13632: ext/fts3/fts3_snippet.c in SQLite before 3.32.0 has a NULL pointer dereference via a crafted matchinfo() query.

Typo3 Team reports:

In case an attacker manages to generate a valid cryptographic message authentication code (HMAC-SHA1) - either by using a different existing vulnerability or in case the internal encryptionKey was exposed - it is possible to retrieve arbitrary files of a TYPO3 installation. This includes the possibility to fetch typo3conf/LocalConfiguration.php which again contains the encryptionKey as well as credentials of the database management system being used. In case a database server is directly accessible either via internet or in a shared hosting network, this allows to completely retrieve, manipulate or delete database contents. This includes creating an administration user account - which can be used to trigger remote code execution by injecting custom extensions.

It has been discovered that an internal verification mechanism can be used to generate arbitrary checksums. This allows to inject arbitrary data having a valid cryptographic message authentication code (HMAC-SHA1) and can lead to various attack chains as described below.

The X.org project reports:

Allocation for pixmap data in AllocatePixmap() does not initialize the memory in xserver, it leads to leak uninitialize heap memory to clients. When the X server runs with elevated privileges.

This flaw can lead to ASLR bypass, which when combined with other flaws (known/unknown) could lead to lead to privilege elevation in the client.

The X.org project reports:

The X Input Method (XIM) client implementation in libX11 has some integer overflows and signed/unsigned comparison issues that can lead to heap corruption when handling malformed messages from an input method.

Python reports:

bpo-41304: Fixes python3x._pth being ignored on Windows, caused by the fix for bpo-29778 (CVE-2020-15801).

bpo-39603: Prevent http header injection by rejecting control characters in http.client.putreques().

KDE Project Security Advisory reports:

KDE Project Security Advisory

Title:	Ark: maliciously crafted archive can install files outside the extraction directory.
Risk Rating:	Important
CVE:	CVE-2020-16116
Versions:	ark <= 20.04.3
Author:	Elvis Angelaccio
Date:	30 July 2020

Overview

A maliciously crafted archive with "../" in the file paths would install files anywhere in the user's home directory upon extraction.

Proof of concept

For testing, an example of malicious archive can be found at https://github.com/jwilk/traversal-archives/releases/download/0/relative2.zip

Impact

Users can unwillingly install files like a modified .bashrc, or a malicious script placed in ~/.config/autostart

Workaround

Users should not use the 'Extract' context menu from the Dolphin file manager. Before extracting a downloaded archive using the Ark GUI, users should inspect it to make sure it doesn't contain entries with "../" in the file path.

Solution

Ark 20.08.0 prevents loading of malicious archives and shows a warning message to the users.

Alternatively, https://invent.kde.org/utilities/ark/-/commit/0df592524fed305d6fbe74ddf8a196bc9ffdb92f can be applied to previous releases.

Credits

Thanks to Dominik Penner for finding and reporting this issue and thanks to Elvis Angelaccio and Albert Astals Cid for fixing it.

RedHat reports:

It was discovered the fix for CVE-2018-19758 was not complete and still allows a read beyond the limits of a buffer in wav_write_header() function in wav.c. A local attacker may use this flaw to make the application crash.

Chrome Releases reports:

This update contains 8 security fixes, including:

• [1105318] High CVE-2020-6537: Type Confusion in V8. Reported by Alphalaab on 2020-07-14
• [1096677] High CVE-2020-6538: Inappropriate implementation in WebView. Reported by Yongke Wang(@Rudykewang) and Aryb1n(@aryb1n) of Tencent Security Xuanwu Lab on 2020-06-18
• [1104061] High CVE-2020-6532: Use after free in SCTP. Reported by Anonymous on 2020-07-09
• [1105635] High CVE-2020-6539: Use after free in CSS. Reported by Oriol Brufau on 2020-07-14
• [1105720] High CVE-2020-6540: Heap buffer overflow in Skia. Reported by Zhen Zhou of NSFOCUS Security Team on 2020-07-15
• [1106773] High CVE-2020-6541: Use after free in WebUSB. Reported by Sergei Glazunov of Google Project Zero on 2020-07-17

Bernhard Miklautz reports:

• Integer overflow due to missing input sanitation in rdpegfx channel
• All FreeRDP clients are affected
• The input rectangles from the server are not checked against local surface coordinates and blindly accepted. A malicious server can send data that will crash the client later on (invalid length arguments to a memcpy)

Jon Siwek of Corelight reports:

This release fixes the following security issues:

• Fix potential DNS analyzer stack overflow
• Fix potential NetbiosSSN analyzer stack overflow

Cacti developers reports:

Multiple fixes for bundled jQuery to prevent code exec (CVE-2020-11022, CVE-2020-11023).

PHPMail contains a escaping bug (CVE-2020-13625).

SQL Injection via color.php in Cacti (CVE-2020-14295).

GitHub Advisory Database:

When a form page type is made available to Wagtail editors through the wagtail.contrib.forms app, and the page template is built using Django's standard form rendering helpers such as form.as_p (as directed in the documentation), any HTML tags used within a form field's help text will be rendered unescaped in the page. Allowing HTML within help text is an intentional design decision by Django; however, as a matter of policy Wagtail does not allow editors to insert arbitrary HTML by default, as this could potentially be used to carry out cross-site scripting attacks, including privilege escalation. This functionality should therefore not have been made available to editor-level users.

The vulnerability is not exploitable by an ordinary site visitor without access to the Wagtail admin.

The Apache Software Foundation reports:

An h2c direct connection did not release the HTTP/1.1 processor after the upgrade to HTTP/2. If a sufficient number of such requests were made, an OutOfMemoryException could occur leading to a denial of service.

The payload length in a WebSocket frame was not correctly validated. Invalid payload lengths could trigger an infinite loop. Multiple requests with invalid payload lengths could lead to a denial of service.

A specially crafted sequence of HTTP/2 requests could trigger high CPU usage for several seconds. If a sufficient number of such requests were made on concurrent HTTP/2 connections, the server could become unresponsive.

Python reports:

bpo-41162:Audit hooks are now cleared later during finalization to avoid missing events.

bpo-29778:Ensure python3.dll is loaded from correct locations when Python is embedded.

Oracle reports:

Vulnerabilities in VirtualBox core can allow users with logon access to the infrastructure where Oracle VM VirtualBox executes to compromise Oracle VM VirtualBox. Successful attacks of these vulnerabilities can result in unauthorized access to critical data, access to all Oracle VM VirtualBox accessible data, unauthorized ability to cause a hang or frequently repeatable crash (complete DOS) or takeover of Oracle VM VirtualBox.

Cary Phillips reports:

openexr 2.5.2 [is a p]atch release with various bug/security and build/install fixes:

• Invalid input could cause a heap-use-after-free error in DeepScanLineInputFile::DeepScanLineInputFile()
• Invalid chunkCount attributes could cause heap buffer overflow in getChunkOffsetTableSize()
• Invalid tiled input file could cause invalid memory access TiledInputFile::TiledInputFile()

Micah Snyder reports:

CVE-2020-3350

Fixed a vulnerability a malicious user could exploit to replace a scan target's directory with a symlink to another path to trick clamscan, clamdscan, or clamonacc into removing or moving a different file (such as a critical system file). The issue would affect users that use the --move or --remove options for clamscan, clamdscan and clamonacc.

CVE-2020-3327

Fixed a vulnerability in the ARJ archive-parsing module in ClamAV 0.102.3 that could cause a denial-of-service (DoS) condition. Improper bounds checking resulted in an out-of-bounds read that could cause a crash. The previous fix for this CVE in version 0.102.3 was incomplete. This fix correctly resolves the issue.

CVE-2020-3481

Fixed a vulnerability in the EGG archive module in ClamAV 0.102.0 - 0.102.3 that could cause a denial-of-service (DoS) condition. Improper error handling could cause a crash due to a NULL pointer dereference. This vulnerability is mitigated for those using the official ClamAV signature databases because the file type signatures in daily.cvd will not enable the EGG archive parser in affected versions.

Jenkins Security Advisory:

Description

(High) SECURITY-1868 / CVE-2020-2220

Stored XSS vulnerability in job build time trend

(High) SECURITY-1901 / CVE-2020-2221

Stored XSS vulnerability in upstream cause

(High) SECURITY-1902 / CVE-2020-2222

Stored XSS vulnerability in 'keep forever' badge icons

(High) SECURITY-1945 / CVE-2020-2223

Stored XSS vulnerability in console links

Chrome Releases reports:

This update contains 38 security fixes, including:

• [1103195] Critical CVE-2020-6510: Heap buffer overflow in background fetch. Reported by Leecraso and Guang Gong of 360 Alpha Lab working with 360 BugCloud on 2020-07-08
• [1074317] High CVE-2020-6511: Side-channel information leakage in content security policy. Reported by Mikhail Oblozhikhin on 2020-04-24
• [1084820] High CVE-2020-6512: Type Confusion in V8. Reported by nocma, leogan, cheneyxu of WeChat Open Platform Security Team on 2020-05-20
• [1091404] High CVE-2020-6513: Heap buffer overflow in PDFium. Reported by Aleksandar Nikolic of Cisco Talos on 2020-06-04
• [1076703] High CVE-2020-6514: Inappropriate implementation in WebRTC. Reported by Natalie Silvanovich of Google Project Zero on 2020-04-30
• [1082755] High CVE-2020-6515: Use after free in tab strip. Reported by DDV_UA on 2020-05-14
• [1092449] High CVE-2020-6516: Policy bypass in CORS. Reported by Yongke Wang(@Rudykewang) and Aryb1n(@aryb1n) of Tencent Security Xuanwu Lab on 2020-06-08
• [1095560] High CVE-2020-6517: Heap buffer overflow in history. Reported by ZeKai Wu (@hellowuzekai) of Tencent Security Xuanwu Lab on 2020-06-16
• [986051] Medium CVE-2020-6518: Use after free in developer tools. Reported by David Erceg on 2019-07-20
• [1064676] Medium CVE-2020-6519: Policy bypass in CSP. Reported by Gal Weizman (@WeizmanGal) of PerimeterX on 2020-03-25
• [1092274] Medium CVE-2020-6520: Heap buffer overflow in Skia. Reported by Zhen Zhou of NSFOCUS Security Team on 2020-06-08
• [1075734] Medium CVE-2020-6521: Side-channel information leakage in autofill. Reported by Xu Lin (University of Illinois at Chicago), Panagiotis Ilia (University of Illinois at Chicago), Jason Polakis (University of Illinois at Chicago) on 2020-04-27
• [1052093] Medium CVE-2020-6522: Inappropriate implementation in external protocol handlers. Reported by Eric Lawrence of Microsoft on 2020-02-13
• [1080481] Medium CVE-2020-6523: Out of bounds write in Skia. Reported by Liu Wei and Wu Zekai of Tencent Security Xuanwu Lab on 2020-05-08
• [1081722] Medium CVE-2020-6524: Heap buffer overflow in WebAudio. Reported by Sung Ta (@Mipu94) of SEFCOM Lab, Arizona State University on 2020-05-12
• [1091670] Medium CVE-2020-6525: Heap buffer overflow in Skia. Reported by Zhen Zhou of NSFOCUS Security Team on 2020-06-05
• [1074340] Low CVE-2020-6526: Inappropriate implementation in iframe sandbox. Reported by Jonathan Kingston on 2020-04-24
• [992698] Low CVE-2020-6527: Insufficient policy enforcement in CSP. Reported by Zhong Zhaochen of andsecurity.cn on 2019-08-10
• [1063690] Low CVE-2020-6528: Incorrect security UI in basic auth. Reported by Rayyan Bijoora on 2020-03-22
• [978779] Low CVE-2020-6529: Inappropriate implementation in WebRTC. Reported by kaustubhvats7 on 2019-06-26
• [1016278] Low CVE-2020-6530: Out of bounds memory access in developer tools. Reported by myvyang on 2019-10-21
• [1042986] Low CVE-2020-6531: Side-channel information leakage in scroll to text. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2020-01-17
• [1069964] Low CVE-2020-6533: Type Confusion in V8. Reported by Avihay Cohen @ SeraphicAlgorithms on 2020-04-11
• [1072412] Low CVE-2020-6534: Heap buffer overflow in WebRTC. Reported by Anonymous on 2020-04-20
• [1073409] Low CVE-2020-6535: Insufficient data validation in WebUI. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2020-04-22
• [1080934] Low CVE-2020-6536: Incorrect security UI in PWAs. Reported by Zhiyang Zeng of Tencent security platform department on 2020-05-09

Oracle reports:

This Critical Patch Update contains 40 new security patches for Oracle MySQL. 6 of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials.

The highest CVSS v3.1 Base Score of vulnerabilities affecting Oracle MySQL is 9.8.

This Pre-Release Announcement provides advance information about the Oracle Critical Patch Update for July 2020, which will be released on Tuesday, July 14, 2020.

NLNetLabs reports:

This release fixes CVE-2020-12662 and CVE-2020-12663.

Bug Fixes:

• CVE-2020-12662 Unbound can be tricked into amplifying an incoming query into a large number of queries directed to a target.
• CVE-2020-12663 Malformed answers from upstream name servers can be used to make Unbound unresponsive.

Problem Description:

The IPV6_2292PKTOPTIONS set handler was missing synchronization, so racing accesses could modify freed memory.

Impact:

A malicious user application could trigger memory corruption, leading to privilege escalation.

The WebKitGTK project reports vulnerabilities:

• CVE-2020-9802: Processing maliciously crafted web content may lead to arbitrary code execution.
• CVE-2020-9803: Processing maliciously crafted web content may lead to arbitrary code execution.
• CVE-2020-9805: Processing maliciously crafted web content may lead to universal cross site scripting.
• CVE-2020-9806: Processing maliciously crafted web content may lead to arbitrary code execution.
• CVE-2020-9807: Processing maliciously crafted web content may lead to arbitrary code execution.
• CVE-2020-9843: Processing maliciously crafted web content may lead to a cross site scripting attack.
• CVE-2020-9850: A remote attacker may be able to cause arbitrary code execution.
• CVE-2020-13753: 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.

Problem Description:

posix_spawnp spawns a new thread with a limited stack allocated on the heap before delegating to execvp for the final execution within that thread.

execvp would previously make unbounded allocations on the stack, directly proportional to the length of the user-controlled PATH environment variable.

Impact:

Long values in the user-controlled PATH environment variable cause posix_spawnp to write beyond the end of stack that was allocated, ultimately overflowing the heap-allocated stack with a direct copy of the value stored in PATH.

mybb Team reports:

High risk: Installer RCE on settings file write

Medium risk: Arbitrary upload paths and Local File Inclusion RCE

Medium risk: XSS via insufficient HTML sanitization of Blog feed and Extend data

Low risk: Open redirect on login

Low risk: SCEditor reflected XSS

kramdown news:

CVE-2020-14001 is addressed to avoid problems when using the {::options /} extension together with the 'template' option.

Manuel Pégourié-Gonnard reports:

The scalar multiplication function in Mbed TLS accepts a random number generator (RNG) as an optional argument and, if provided, uses it to protect against some attacks.

It is the caller's responsibility to provide a RNG if protection against side-channel attacks is desired; however two groups of functions in Mbed TLS itself fail to pass a RNG:

1. mbedtls_pk_parse_key() and mbedtls_pk_parse_keyfile()
2. mbedtls_ecp_check_pub_priv() and mbedtls_pk_check_pair()

When those functions are called, scalar multiplication is computed without randomisation, a number of old and new attacks apply, allowing a powerful local attacker to fully recover the private key.

Gitlab reports:

Workhorse bypass allows files in /tmp to be read via Maven Repository APIs

Python reports:

The AbstractBasicAuthHandler class of the urllib.request module uses an inefficient regular expression which can be exploited by an attacker to cause a denial of service. Fix the regex to prevent the catastrophic backtracking. Vulnerability reported by Ben Caller and Matt Schwager.

Disallow control characters in hostnames in http.client, addressing CVE-2019-18348. Such potentially malicious header injection URLs now cause a InvalidURL to be raised.

Disallow CR or LF in email.headerregistry.Address arguments to guard against header injection attacks.

Anydesk reports:

AnyDesk before 5.5.3 on Linux and FreeBSD has a format string vulnerability that can be exploited for remote code execution.

GitHub Security Lab reports:

D-Bus has a file descriptor leak, which can lead to denial of service when the dbus-daemon runs out of file descriptors. An unprivileged local attacker can use this to attack the system dbus-daemon, leading to denial of service for all users of the machine.

Matrix developers report:

Due to the two security issues highlighted below, server administrators are encouraged to update Synapse. We are not aware of these vulnerabilities being exploited in the wild.

• A malicious homeserver could force Synapse to reset the state in a room to a small subset of the correct state. This affects all Synapse deployments which federate with untrusted servers.
• HTML pages served via Synapse were vulnerable to clickjacking attacks. This predominantly affects homeservers with single-sign-on enabled, but all server administrators are encouraged to upgrade.

Gitlab reports:

Missing Permission Check on Time Tracking

Cross-Site Scripting in PyPi Files API

Insecure Authorization Check on Private Project Security Dashboard

Cross-Site Scripting in References

Cross-Site Scripting in Group Names

Cross-Site Scripting in Blob Viewer

Cross-Site Scripting in Error Tracking

Insecure Authorisation Check on Creation and Deletion of Deploy Tokens

User Name Format Restiction Bypass

Denial of Service in Issue Comments

Cross-Site Scripting in Wiki Pages

Private Merge Request Updates Leaked via Todos

Private User Activity Leaked via API

Cross-Site Scripting in Bitbucket Import Feature

Github Project Restriction Bypass

Update PCRE Dependency

Update Kaminari Gem

Cross-Site Scripting in User Profile

Update Xterm.js

PowerDNS Team reports:

CVE-2020-14196: An issue has been found in PowerDNS Recursor where the ACL applied to the internal web server via webserver-allow-from is not properly enforced, allowing a remote attacker to send HTTP queries to the internal web server, bypassing the restriction. In the default configuration the API webserver is not enabled. Only installations using a non-default value for webserver and webserver-address are affected.

Bryan Call reports:

ATS is vulnerable to certain types of HTTP/2 HEADERS frames that can cause the server to allocate a large amount of memory and spin the thread.

The Samba Team reports:

Four vulnerabilities were fixed in samba:

• CVE-2020-10730: NULL pointer de-reference and use-after-free in Samba AD DC LDAP Server with ASQ, VLV and paged_results
• CVE-2020-10745: Parsing and packing of NBT and DNS packets can consume excessive CPU in the AD DC (only)
• CVE-2020-10760: LDAP Use-after-free in Samba AD DC Global Catalog with paged_results and VLV
• CVE-2020-14303: Empty UDP packet DoS in Samba AD DC nbtd

Felix Dörre reports:

The issue is that STUN/TURN response buffer is not initialized properly. (CWE 665) This is a leak of information between different client connections. One client (an attacker) could use their connection to intelligently query coturn to get interesting bytes in the padding bytes from the connection of another client.

Drupal Security Team reports:

The Drupal core Form API does not properly handle certain form input from cross-site requests, which can lead to other vulnerabilities.

Ashley Newson reports:

The xrdp-sesman service can be crashed by connecting over port 3350 and supplying a malicious payload. Once the xrdp-sesman process is dead, an unprivileged attacker on the server could then proceed to start their own imposter sesman service listening on port 3350.

reports:

Improper serialization of MongoDB Server's internal authorization state permits a user with valid credentials to bypass IP source address protection mechanisms following administrative action.

Credit

Discovered by Tony Yesudas.

Two vulnerabilities were fixed in the upstream repository:

• The bark_noise_hybridmp function allows remote attackers to cause a denial of service (out-of-bounds access and application crash) or possibly have unspecified other impact via a crafted file.
• mapping0_forward does not validate the number of channels, which allows remote attackers to cause a denial of service (heap-based buffer overflow or over-read) or possibly have unspecified other impact via a crafted file.

Simon Tatham reports:

[Release 0.74] fixes the following security issues:

• New configuration option to disable PuTTY's default policy of changing its host key algorithm preferences to prefer keys it already knows. (There is a theoretical information leak in this policy.) [CVE-2020-14002]
• In some situations an SSH server could cause PuTTY to access freed mdmory by pretending to accept an SSH key and then refusing the actual signature. It can only happen if you're using an SSH agent.

mutt 1.14.4 updates:

CVE-2020-14954 - Machine-in-the-middle response injection attack when using STARTTLS with IMAP, POP3, and SMTP

Apple reports:

• CVE-2019-8842: The ippReadIO function may under-read an extension.
• CVE-2020-3898: The ppdOpen function did not handle invalid UI constraint. ppdcSource::get_resolution function did not handle invalid resolution strings. An application may be able to gain elevated privileges.

mutt 1.14.3 updates:

CVE-2020-14093 - IMAP fcc/postpone man-in-the-middle attack via a PREAUTH response.

Chrome Releases reports:

This update includes 2 security fixes, including:

• [1092308] High CVE-2020-6509: Use after free in extensions. Reported by Anonymous on 2020-06-08

curl security problems:

CVE-2020-8169: Partial password leak over DNS on HTTP redirect

libcurl can be tricked to prepend a part of the password to the host name before it resolves it, potentially leaking the partial password over the network and to the DNS server(s).

libcurl can be given a username and password for HTTP authentication when requesting an HTTP resource - used for HTTP Authentication such as Basic, Digest, NTLM and similar. The credentials are set, either together with CURLOPT_USERPWD or separately with CURLOPT_USERNAME and CURLOPT_PASSWORD. Important detail: these strings are given to libcurl as plain C strings and they are not supposed to be URL encoded.

In addition, libcurl also allows the credentials to be set in the URL, using the standard RFC 3986 format: http://user:password@host/path. In this case, the name and password are URL encoded as that's how they appear in URLs.

If the options are set, they override the credentials set in the URL.

Internally, this is handled by storing the credentials in the "URL object" so that there is only a single set of credentials stored associated with this single URL.

When libcurl handles a relative redirect (as opposed to an absolute URL redirect) for an HTTP transfer, the server is only sending a new path to the client and that path is applied on to the existing URL. That "applying" of the relative path on top of an absolute URL is done by libcurl first generating a full absolute URL out of all the components it has, then it applies the redirect and finally it deconstructs the URL again into its separate components.

This security vulnerability originates in the fact that curl did not correctly URL encode the credential data when set using one of the curl_easy_setopt options described above. This made curl generate a badly formatted full URL when it would do a redirect and the final re-parsing of the URL would then go bad and wrongly consider a part of the password field to belong to the host name.

The wrong host name would then be used in a name resolve lookup, potentially leaking the host name + partial password in clear text over the network (if plain DNS was used) and in particular to the used DNS server(s).

CVE-2020-8177: curl overwrite local file with -J

curl can be tricked by a malicious server to overwrite a local file when using -J (--remote-header-name) and -i (--include) in the same command line.

The command line tool offers the -J option that saves a remote file using the file name present in the Content-Disposition: response header. curl then refuses to overwrite an existing local file using the same name, if one already exists in the current directory.

The -J flag is designed to save a response body, and so it doesn't work together with -i and there's logic that forbids it. However, the check is flawed and doesn't properly check for when the options are used in the reversed order: first using -J and then -i were mistakenly accepted.

The result of this mistake was that incoming HTTP headers could overwrite a local file if one existed, as the check to avoid the local file was done first when body data was received, and due to the mistake mentioned above, it could already have received and saved headers by that time.

The saved file would only get response headers added to it, as it would abort the saving when the first body byte arrives. A malicious server could however still be made to send back virtually anything as headers and curl would save them like this, until the first CRLF-CRLF sequence appears.

(Also note that -J needs to be used in combination with -O to have any effect.)

Joel Esler reports:

• CVE-2018-15378:
  • Vulnerability in ClamAV's MEW unpacking feature that could allow an unauthenticated, remote attacker to cause a denial of service (DoS) condition on an affected device.
  • Reported by Secunia Research at Flexera.
• Fix for a 2-byte buffer over-read bug in ClamAV&s PDF parsing code.
  • Reported by Alex Gaynor.
• CVE-2018-14680:
  • An issue was discovered in mspack/chmd.c in libmspack before 0.7alpha. It does not reject blank CHM filenames.
• CVE-2018-14681:
  • An issue was discovered in kwajd_read_headers in mspack/kwajd.c in libmspack before 0.7alpha. Bad KWAJ file header extensions could cause a one or two byte overwrite.
• CVE-2018-14682:
  • An issue was discovered in mspack/chmd.c in libmspack before 0.7alpha. There is an off-by-one error in the TOLOWER() macro for CHM decompression. Additionally, 0.100.2 reverted 0.100.1's patch for CVE-2018-14679, and applied libmspack's version of the fix in its place.

Apple reports:

• CVE-2019-8842: The ippReadIO function may under-read an extension.
• CVE-2020-3898: The ppdOpen function did not handle invalid UI constraint. ppdcSource::get_resolution function did not handle invalid resolution strings. An application may be able to gain elevated privileges.

CVE list:

GNU libidn2 before 2.2.0 fails to perform the roundtrip checks specified in RFC3490 Section 4.2 when converting A-labels to U-labels. This makes it possible in some circumstances for one domain to impersonate another. By creating a malicious domain that matches a target domain except for the inclusion of certain punycoded Unicode characters (that would be discarded when converted first to a Unicode label and then back to an ASCII label), arbitrary domains can be impersonated.

Marina Glancy reports:

• MSA-17-0001: System file inclusion when adding own preset file in Boost theme

• MSA-17-0002: Incorrect sanitation of attributes in forums

• MSA-17-0003: PHPMailer vulnerability in no-reply address

• MSA-17-0004: XSS in assignment submission page

Ruby on Rails blog:

Rails 6.0.3.2 has been released! This version of Rails contains an important security patch, and you should upgrade! The release contains only one patch that addresses CVE-2020-8185.

ISC reports:

An assertion check in BIND (that is meant to prevent going beyond the end of a buffer when processing incoming data) can be incorrectly triggered by a large response during zone transfer.

ISC reports:

The asterisk character ("*") is allowed in DNS zone files, where it is most commonly present as a wildcard at a terminal node of the Domain Name System graph. However, the RFCs do not require and BIND does not enforce that an asterisk character be present only at a terminal node.

A problem can occur when an asterisk is present in an empty non-terminal location within the DNS graph. If such a node exists, after a series of queries, named can reach an inconsistent state that results in the failure of an assertion check in rbtdb.c, followed by the program exiting due to the assertion failure.

lynis update:

This release resolves two security issues

• CVE-2020-13882 - Discovered by Sander Bos, code submission by Katarina Durechova
• CVE-2019-13033 - Discovered by Sander Bos

Thomas Guillem reports:

A heap-based buffer overflow in the hxxx_AnnexB_to_xVC function in modules/packetizer/hxxx_nal.c in VideoLAN VLC media player before 3.0.11 allows remote attackers to cause a denial of service (application crash) or execute arbitrary code via a crafted H.264 Annex-B video (.avi for example) file.

Ben Caller and Matt Schwager reports:

Python 2.7 through 2.7.17, 3.5 through 3.5.9, 3.6 through 3.6.10, 3.7 through 3.7.6, and 3.8 through 3.8.1 allows an HTTP server to conduct Regular Expression Denial of Service (ReDoS) attacks against a client because of urllib.request.AbstractBasicAuthHandler catastrophic backtracking.

Python reports:

An issue was discovered in urllib2 in Python 2.x through 2.7.17 and urllib in Python 3.x through 3.8.0. CRLF injection is possible if the attacker controls a url parameter, as demonstrated by the first argument to urllib.request.urlopen with \r\n (specifically in the host component of a URL) followed by an HTTP header.

Node.js reports:

Updates are now available for all supported Node.js release lines for the following issues.

TLS session reuse can lead to host certificate verification bypass (High) (CVE-2020-8172)

The 'session' event could be emitted before the 'secureConnect' event. It should not be, because the connection may fail to be authorized. If it was saved an authorized connection could be established later with the session ticket. Note that the https agent caches sessions, so is vulnerable to this.

The 'session' event will now only be emitted after the 'secureConnect' event, and only for authorized connections.

HTTP/2 Large Settings Frame DoS (Low) (CVE-2020-11080)

Receiving unreasonably large HTTP/2 SETTINGS frames can consume 100% CPU to process all the settings, blocking all other activities until complete.

The HTTP/2 session frame is limited to 32 settings by default. This can be configured if necessary using the maxSettings option.

napi_get_value_string_*() allows various kinds of memory corruption (High) (CVE-2020-8174)

Calling napi_get_value_string_latin1(), napi_get_value_string_utf8(), or napi_get_value_string_utf16() with a non-NULL buf, and a bufsize of 0 will cause the entire string value to be written to buf, probably overrunning the length of the buffer.

A exploit has not been reported and it may be difficult but the following is suggested:

• All users of LTS Node.js versions should update to the versions announced in this security post. This will address the issue for any non pre-built add-on.
• Maintainers who support EOL Node.js versions and/or build against a version of Node.js that did not support N-API internally should update to use the new versions of node-addon-api 1.x and 2.x that will be released soon after this announcement.

ICU-20958 Prevent SEGV_MAPERR in append (High) (CVE-2020-10531)

An issue was discovered in International Components for Unicode (ICU) for C/C++ through 66.1. An integer overflow, leading to a heap-based buffer overflow, exists in the UnicodeString::doAppend() function in common/unistr.cpp.

Fix was applied to 10.x in an abundance of caution, even though there is no known way to trigger the overflow in 10.x.

LibreOffice reports:

Two flaws were found in LibreOffice:

• CVE-2020-12802: remote graphics contained in docx format retrieved in 'stealth mode'
• CVE-2020-12803: XForms submissions could overwrite local files

fklassen on Github reports:

This release fixes the following security issues:

• memory access in do_checksum()
• NULL pointer dereference get_layer4_v6()
• NULL pointer dereference get_ipv6_l4proto()

Mitre reports:

ZNC 1.8.0 up to 1.8.1-rc1 allows attackers to trigger an application crash (with a NULL pointer dereference) if echo-message is not enabled and there is no network.

NPM reports:

Global node_modules Binary Overwrite

Symlink reference outside of node_modules

Arbitrary File Write

Malvineous on Github reports:

This release fixes the following security issues:

• buffer overflow in .bmf
• buffer overflow in .dtm
• buffer overflow in .mkj
• buffer overflow in .a2m
• buffer overflow in .rad
• buffer overflow in .mtk
• double free and OOB reads in .u6m

Jon Siwek of Corelight reports:

This release fixes the following security issues:

• Fix potential stack overflow in NVT analyzer
• Fix NVT analyzer memory leak from multiple telnet authn name options
• Fix multiple content-transfer-encoding headers causing a memory leak
• Fix potential leak of Analyzers added to tree during Analyzer::Done
• Prevent IP fragment reassembly on packets without minimal IP header

Adobe reports:

• This update resolves a use-after-free vulnerability that could lead to arbitrary code execution (CVE-2020-9633).

Problem Description:

If the push/pop level of the USB HID state is not restored within the processing of the same HID item, an invalid memory location may be used for subsequent HID item processing.

Impact:

An attacker with physical access to a USB port may be able to use a specially crafted USB device to gain kernel or user-space code execution.

Chrome Releases reports:

This update includes 5 security fixes. Below, we highlight fixes that were contributed by external researchers.

• [1082105] High CVE-2020-6493: Use after free in WebAuthentication. Reported by Anonymous on 2020-05-13
• [1083972] High CVE-2020-6494: Incorrect security UI in payments. Reported by Juho Nurminen on 2020-05-18
• [1072116] High CVE-2020-6495: Insufficient policy enforcement in developer tools. Reported by David Erceg on 2020-04-18
• [1085990] High CVE-2020-6496: Use after free in payments. Reported by Khalil Zhani on 2020-05-24

Gitlab reports:

CI Token Access Control

Django security release reports:

CVE-2020-13254: Potential data leakage via malformed memcached keys

In cases where a memcached backend does not perform key validation, passing malformed cache keys could result in a key collision, and potential data leakage. In order to avoid this vulnerability, key validation is added to the memcached cache backends.

CVE-2020-13596: Possible XSS via admin ForeignKeyRawIdWidget

Query parameters for the admin ForeignKeyRawIdWidget were not properly URL encoded, posing an XSS attack vector. ForeignKeyRawIdWidget now ensures query parameters are correctly URL encoded.

The GnuTLS project reports:

It was found that GnuTLS 3.6.4 introduced a regression in the TLS protocol implementation. This caused the TLS server to not securely construct a session ticket encryption key considering the application supplied secret, allowing a MitM attacker to bypass authentication in TLS 1.3 and recover previous conversations in TLS 1.2.

nghttp2 security advisories:

The overly large HTTP/2 SETTINGS frame payload causes denial of service.

The proof of concept attack involves a malicious client constructing a SETTINGS frame with a length of 14,400 bytes (2400 individual settings entries) over and over again. The attack causes the CPU to spike at 100%.

Changelog:

Remove a ReDoS vulnerability in the header parser (CVE-2020-7663)

The Gitea Team reports for release 1.11.6:

• Fix missing authorization check on pull for public repos of private/limited org (#11656) (#11683)
• Use session for retrieving org teams (#11438) (#11439)

PowerDNS Team reports:

CVE-2020-10995: An issue in the DNS protocol has been found that allow malicious parties to use recursive DNS services to attack third party authoritative name servers. The attack uses a crafted reply by an authoritative name server to amplify the resulting traffic between the recursive and other authoritative name servers. Both types of service can suffer degraded performance as an effect.

CVE-2020-12244: An issue has been found in PowerDNS Recursor 4.1.0 through 4.3.0 where records in the answer section of a NXDOMAIN response lacking an SOA were not properly validated in SyncRes::processAnswer. This would allow an attacker in position of man-in-the-middle to send a NXDOMAIN answer for a name that does exist, bypassing DNSSEC validation.

CVE-2020-10030: An issue has been found in PowerDNS Authoritative Server allowing an attacker with enough privileges to change the system's hostname to cause disclosure of uninitialized memory content via a stack-based out-of-bounds read. It only occurs on systems where gethostname() does not null-terminate the returned string if the hostname is larger than the supplied buffer. Linux systems are not affected because the buffer is always large enough. OpenBSD systems are not affected because the returned hostname is always null-terminated. Under some conditions this issue can lead to the writing of one null-byte out-of-bounds on the stack, causing a denial of service or possibly arbitrary code execution.

The Sane Project reports:

epson2: fixes CVE-2020-12867 (GHSL-2020-075) and several memory management issues found while addressing that CVE

epsonds: addresses out-of-bound memory access issues to fix CVE-2020-12862 (GHSL-2020-082) and CVE-2020-12863 (GHSL-2020-083), addresses a buffer overflow fixing CVE-2020-12865 (GHSL-2020-084) and disables network autodiscovery to mitigate CVE-2020-12866 (GHSL-2020-079), CVE-2020-12861 (GHSL-2020-080) and CVE-2020-12864 (GHSL-2020-081). Note that this backend does not support network scanners to begin with.

magicolor: fixes a floating point exception and uninitialized data read

fixes an overflow in sanei_tcp_read()

Kaminari Security Advisories:

There was a vulnerability in versions of Kaminari that would allow an attacker to inject arbitrary code into pages with pagination links.

The 1.2.1 gem including the patch has already been released.

All past released versions are affected by this vulnerability.

The FreeRDP changelog reports 14 CVEs addressed after 2.0.0-rc4

Gitlab reports:

User Email Verification Bypass

OAuth Flow Missing Email Verification Checks

Notification Email Verification Bypass

Undisclosed Vulnerability on a Third-Party Rendering Engine

Group Sign-Up Restriction Bypass

Mirror Project Owner Impersonation

Missing Permission Check on Fork Relation Creation

Cross-Site Scripting in Repository Files API

Kubernetes Cluster Token Disclosure

Object Storage File Enumeration

Insecure Authorization Check on Project Deploy Keys

Cross-Site Scripting on Metrics Dashboard

Denial of Service on Custom Dashboards

Client-Side Code Injection through Mermaid Markup

Cross-Site Scripting on Static Site Editor

Disclosure of Amazon EKS Credentials

Denial of Service on Workhorse

A vulnerability has been discovered in Sympa web interface by which attacker can execute arbitrary code with root privileges. Sympa uses two sorts of setuid wrappers:

• FastCGI wrappers
• newaliases wrapper

The FastCGI wrappers wwsympa-wrapper.fcgi and sympa_soap_server-wrapper.fcgi were used to make the web interface running under privileges of a dedicated user.

The newaliases wrapper (sympa_newaliases-wrapper) allows Sympa to update the alias database with root privileges.

Since these setuid wrappers did not clear environment variables, if environment variables like PERL5LIB were injected, forged code might be loaded and executed under privileges of setuid-ed users.

Google Chrome Releases reports:

This release includes 38 security fixes, including CVEs CVE-2020-6465 through CVE-2020-6491.

Piwigo reports:

Piwigo 2.10.1 is affected by stored XSS via the Group Name Field to the group_list page.

The Apache Software Foundation reports:

Under certain circumstances an attacker will be able to trigger remote code execution via deserialization of the file under their control

Javier Moreno discovered a vulnerability in Sympa web interface that can cause denial of service (DoS) attack.

By submitting requests with malformed parameters, this flaw allows to create junk files in Sympa's directory for temporary files. And particularly by tampering token to prevent CSRF, it allows to originate exessive notification messages to listmasters.

Drupal Security Team reports:

The jQuery project released version 3.5.0, and as part of that, disclosed two security vulnerabilities that affect all prior versions. As mentioned in the jQuery blog, both are: ... Security issues in jQuerys DOM manipulation methods, as in .html(), .append(), and the others. Security advisories for both of these issues have been published on GitHub.

Drupal 7 has an Open Redirect vulnerability. For example, a user could be tricked into visiting a specially crafted link which would redirect them to an arbitrary external URL. The vulnerability is caused by insufficient validation of the destination query parameter in the drupal_goto() function.

Zabbix reports:

Fixed security vulnerability cve-2020-11800 (remote code execution). (ZBX-17600)

Ruby on Rails blog:

Hi everyone! Rails 5.2.4.3 and 6.0.3.1 have been released! These releases contain important security fixes, so please upgrade when you can.

Both releases contain the following fixes:

CVE-2020-8162: Circumvention of file size limits in ActiveStorage

CVE-2020-8164: Possible Strong Parameters Bypass in ActionPack

CVE-2020-8165: Potentially unintended unmarshalling of user-provided objects in MemCacheStore and RedisCacheStore

CVE-2020-8166: Ability to forge per-form CSRF tokens given a global CSRF token

CVE-2020-8167: CSRF Vulnerability in rails-ujs

Aki Tuomi reports:

Vulnerability Details: Sending malformed NOOP command causes crash in submission, submission-login or lmtp service. Risk: Remote attacker can keep submission-login service down, causing denial of service attack. For lmtp the risk is neglible, as lmtp is usually behind a trusted MTA. Steps to reproduce: Send ``NOOP EE"FY`` to submission port, or similarly malformed command.

Vulnerability Details: Sending command followed by sufficient number of newlines triggers a use-after-free bug that might crash submission-login, submission or lmtp service. Risk: Remote attacker can keep submission-login service down, causing denial of service attack. For lmtp the risk is neglible, as lmtp is usually behind a trusted MTA. Steps to reproduce: This can be currently reproduced with ASAN or Valgrind. Reliable way to crash has not yet been discovered.

Vulnerability Details: Sending mail with empty quoted localpart causes submission or lmtp component to crash. Risk: Malicious actor can cause denial of service to mail delivery by repeatedly sending mails with bad sender or recipient address. Steps to reproduce: Send mail with envelope sender or recipient as <""@example.org>. Workaround: For submission there is no workaround, but triggering the bug requires valid credentials. For lmtp, one can implement sufficient filtering on MTA level to prevent mails with such addresses from ending up in LMTP delivery.

Tobias Stöckmann reports:

I have discovered a way to trigger an out of boundary write while parsing a huge json file through a malicious input source. It can be triggered if an attacker has control over the input stream or if a huge load during filesystem operations can be triggered.

Oracle reports:

This Critical Patch Update contains 45 new security patches for Oracle MySQL. 9 of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials.

MariaDB reports 4 of these vulnerabilities exist in their software

F-Secure reports:

CVE-2020-11651 - Authentication bypass vulnerabilities

The ClearFuncs class processes unauthenticated requests and unintentionally exposes the _send_pub() method, which can be used to queue messages directly on the master publish server. Such messages can be used to trigger minions to run arbitrary commands as root.

The ClearFuncs class also exposes the method _prep_auth_info(), which returns the "root key" used to authenticate commands from the local root user on the master server. This "root key" can then be used to remotely call administrative commands on the master server. This unintentional exposure provides a remote un-authenticated attacker with root-equivalent access to the salt master.

CVE-2020-11652 - Directory traversal vulnerabilities

The wheel module contains commands used to read and write files under specific directory paths. The inputs to these functions are concatenated with the target directory and the resulting path is not canonicalized, leading to an escape of the intended path restriction.

The get_token() method of the salt.tokens.localfs class (which is exposed to unauthenticated requests by the ClearFuncs class) fails to sanitize the token input parameter which is then used as a filename, allowing insertion of ".." path elements and thus reading of files outside of the intended directory. The only restriction is that the file has to be deserializable by salt.payload.Serial.loads().

Ruby on Rails blog:

Due to an unfortunate oversight, Rails 4.2.11.2 has a missing constant error. To address this Rails 4.2.11.3 has been released.

The original announcement for CVE-2020-8163 has a follow-up message with an updated patch if you?re unable to use the gems.

Micah Snyder reports:

CVE-2020-3327: Fixed a vulnerability in the ARJ archive-parsing module in ClamAV 0.102.2 that could cause a denial-of-service condition. Improper bounds checking of an unsigned variable results in an out-of-bounds read which causes a crash. Special thanks to Daehui Chang and Fady Othman for helping identify the ARJ parsing vulnerability.

CVE-2020-3341: Fixed a vulnerability in the PDF-parsing module in ClamAV 0.101 - 0.102.2 that could cause a denial-of-service condition. Improper size checking of a buffer used to initialize AES decryption routines results in an out-of-bounds read, which may cause a crash. OSS-Fuzz discovered this vulnerability.

Typo3 News:

CVE-2020-11063: TYPO3-CORE-SA-2020-001: Information Disclosure in Password Reset

It has been discovered that time-based attacks can be used with the password reset functionality for backend users. This allows an attacker to verify whether a backend user account with a given email address exists or not.

CVE-2020-11064: TYPO3-CORE-SA-2020-002: Cross-Site Scripting in Form Engine

It has been discovered that HTML placeholder attributes containing data of other database records are vulnerable to cross-site scripting. A valid backend user account is needed to exploit this vulnerability.

CVE-2020-11065: TYPO3-CORE-SA-2020-003: Cross-Site Scripting in Link Handling

It has been discovered that link tags generated by typolink functionality are vulnerable to cross-site scripting - properties being assigned as HTML attributes have not been parsed correctly.

CVE-2020-11066: TYPO3-CORE-SA-2020-004: Class destructors causing side-effects when being unserialized

Calling unserialize() on malicious user-submitted content can result in the following scenarios:

- trigger deletion of arbitrary directory in file system (if writable for web server)

- trigger message submission via email using identity of web site (mail relay)

Another insecure deserialization vulnerability is required to actually exploit mentioned aspects.

CVE-2020-11067: TYPO3-CORE-SA-2020-005: Insecure Deserialization in Backend User Settings

It has been discovered that backend user settings (in $BE_USER->uc) are vulnerable to insecure deserialization. In combination with vulnerabilities of 3rd party components this can lead to remote code execution. A valid backend user account is needed to exploit this vulnerability.

CVE-2020-11069: TYPO3-CORE-SA-2020-006: Same-Site Request Forgery to Backend User Interface

It has been discovered that the backend user interface and install tool are vulnerable to same-site request forgery. A backend user can be tricked into interacting with a malicious resource an attacker previously managed to upload to the web server - scripts are then executed with the privileges of the victims? user session.

In a worst case scenario new admin users can be created which can directly be used by an attacker. The vulnerability is basically a cross-site request forgery (CSRF) triggered by a cross-site scripting vulnerability (XSS) - but happens on the same target host - thus, it? actually a same-site request forgery (SSRF).

Malicious payload such as HTML containing JavaScript might be provided by either an authenticated backend user or by a non-authenticated user using a 3rd party extension - e.g. file upload in a contact form with knowing the target location.

The attacked victim requires an active and valid backend or install tool user session at the time of the attack to be successful.

Problem Description:

Requests to create cryptography sessions using a MAC did not validate the user-supplied MAC key length. The cryptodev module allocates a buffer whose size is this user-suppled length.

Impact:

An unprivileged process can trigger a kernel panic.

Problem Description:

The SCTP layer does improper checking when an application tries to update a shared key. Therefore an unprivileged local user can trigger a use-after- free situation, for example by specific sequences of updating shared keys and closing the SCTP association.

Impact:

Triggering the use-after-free situation may result in unintended kernel behaviour including a kernel panic.

Problem Description:

libalias(3) packet handlers do not properly validate the packet length before accessing the protocol headers. As a result, if a libalias(3) module does not properly validate the packet length before accessing the protocol header, it is possible for an out of bound read or write condition to occur.

Impact:

A malicious attacker could send specially constructed packets that exploit the lack of validation allowing the attacker to read or write memory either from the kernel (for the in-kernel NAT implementation) or from the process space for natd (for the userspace implementation).

Problem Description:

The FTP packet handler in libalias incorrectly calculates some packet lengths. This may result in disclosing small amounts of memory from the kernel (for the in-kernel NAT implementation) or from the process space for natd (for the userspace implementation).

Impact:

A malicious attacker could send specially constructed packets that exploit the erroneous calculation allowing the attacker to disclose small amount of memory either from the kernel (for the in-kernel NAT implementation) or from the process space for natd (for the userspace implementation).

Problem Description:

A race condition permitted a data structure in the kernel to be used after it was freed by the cryptodev module.

Impact:

An unprivileged process can overwrite arbitrary kernel memory.

Qutebrowser developers report:

After a certificate error was overridden by the user, qutebrowser displays the URL as yellow (colors.statusbar.url.warn.fg). However, when the affected website was subsequently loaded again, the URL was mistakenly displayed as green (colors.statusbar.url.success_https). While the user already has seen a certificate error prompt at this point (or set content.ssl_strict to false which is not recommended), this could still provide a false sense of security.

MITRE Corporation reports:

inc/user.class.php in GLPI before 9.4.3 allows XSS via a user picture.

Mark Sapiro reports:

A content injection vulnerability via the options login page has been discovered and reported by Vishal Singh.

An issue similar to CVE-2018-13796 exists at different endpoint & param. It can lead to a phishing attack.

(added 2020-05-07) This is essentially the same as https://bugs.launchpad.net/mailman/+bug/1873722 except the vector is the private archive login page and the attack only succeeds if the list's roster visibility (private_roster) setting is 'Anyone'.

Jon Siwek of Corelight reports:

This release fixes the following security issues:

• Fix buffer over-read in Ident analyzer
• Fix SSL scripting error leading to uninitialized field access and memory leak
• Fix POP3 analyzer global buffer over-read
• Fix potential stack overflows due to use of Variable-Length-Arrays

Wagtail release notes:

CVE-2020-11037: Potential timing attack on password-protected private pages

This release addresses a potential timing attack on pages or documents that have been protected with a shared password through Wagtail's "Privacy" controls. This password check is performed through a character-by-character string comparison, and so an attacker who is able to measure the time taken by this check to a high degree of accuracy could potentially use timing differences to gain knowledge of the password. (This is understood to be feasible on a local network, but not on the public internet.)

Cacti developer reports:

Lack of escaping of color items can lead to XSS exposure.

Webin security lab - dbapp security Ltd reports:

The TagLib::Ogg::FLAC::File::scan function in oggflacfile.cpp in TagLib 1.11.1 allows remote attackers to cause information disclosure (heap-based buffer over-read) via a crafted audio file.

Gitlab reports:

Path Traversal in NuGet Package Registry

Workhorse Bypass Leads to File Disclosure

OAuth Application Client Secrets Revealed

Code Owners Approval Rules Are Not Updated for Existing Merge Requests When Source Branch Changes

Code Owners Protection Not Enforced from Web UI

Repository Mirror Passwords Exposed To Maintainers

Admin Audit Log Page Denial of Service

Private Project ID Revealed Through Group API

Elasticsearch Credentials Logged to ELK

GitHub Personal Access Token Exposed on Integrations Page

Update Nokogiri dependency

Update OpenSSL Dependency

Update git

The Samba Team reports:

CVE-2020-10700

A client combining the 'ASQ' and 'Paged Results' LDAP controls can cause a use-after-free in Samba's AD DC LDAP server.

CVE-2020-10704

A deeply nested filter in an un-authenticated LDAP search can exhaust the LDAP server's stack memory causing a SIGSEGV.

VideoLAN reports:

Details

A remote user could:

• Create a specifically crafted image file that could trigger an out of bounds read
• Send a specifically crafter request to the microdns service discovery, potentially triggering various memory management issues

Impact

If successful, a malicious third party could trigger either a crash of VLC or an arbitratry code execution with the privileges of the target user.

While these issues in themselves are most likely to just crash the player, we can't exclude that they could be combined to leak user informations or remotely execute code. ASLR and DEP help reduce the likelyness of code execution, but may be bypassed.

We have not seen exploits performing code execution through these vulnerabilities

CVE-2019-19721 affects VLC 3.0.8 and earlier, and only reads 1 byte out of bound

Riccardo Schirone (https://github.com/ret2libc) reports:

In FullLoader python/object/new constructor, implemented by construct_python_object_apply, has support for setting the state of a deserialized instance through the set_python_instance_state method. After setting the state, some operations are performed on the instance to complete its initialization, however it is possible for an attacker to set the instance' state in such a way that arbitrary code is executed by the FullLoader.

This patch tries to block such attacks in FullLoader by preventing set_python_instance_state from setting arbitrar properties. It implements a blacklist that includes extend method (called by construct_python_object_apply) and all special methods (e.g. __set__, __setitem__, etc.).

Users who need special attributes being set in the state of a deserialized object can still do it through the UnsafeLoader, which however should not be used on untrusted input. Additionally, they can subclass FullLoader and redefine state_blacklist_regexp to include the additional attributes they need, passing the subclassed loader to yaml.load.

Howard Chu reports:

nested filters leads to stack overflow

Bleach developers reports:

bleach.clean behavior parsing style attributes could result in a regular expression denial of service (ReDoS).

Calls to bleach.clean with an allowed tag with an allowed style attribute are vulnerable to ReDoS. For example, bleach.clean(..., attributes={'a': ['style']}).

Oracle reports:

This Critical Patch Update contains 45 new security patches for Oracle MySQL. 9 of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials.

Nextcloud reports:

XSS in Files PDF viewer (NC-SA-2020-019)

Missing ownership check on remote wipe endpoint (NC-SA-2020-018)

Problem Description:

Server or client applications that call the SSL_check_chain() function during or after a TLS 1.3 handshake may crash due to a NULL pointer dereference as a result of incorrect handling of the "signature_algorithms_cert" TLS extension. The crash occurs if an invalid or unrecognized signature algorithm is received from the peer.

Impact:

A malicious peer could exploit the NULL pointer dereference crash, causing a denial of service attack.

git security advisory reports:

Git uses external "credential helper" programs to store and retrieve passwords or other credentials from secure storage provided by the operating system. Specially-crafted URLs that are considered illegal as of the recently published Git versions can cause Git to send a "blank" pattern to helpers, missing hostname and protocol fields. Many helpers will interpret this as matching any URL, and will return some unspecified stored password, leaking the password to an attacker's server.

Wagtail release notes:

CVE-2020-11001: Possible XSS attack via page revision comparison view

This release addresses a cross-site scripting (XSS) vulnerability on the page revision comparison view within the Wagtail admin interface. A user with a limited-permission editor account for the Wagtail admin could potentially craft a page revision history that, when viewed by a user with higher privileges, could perform actions with that user credentials. The vulnerability is not exploitable by an ordinary site visitor without access to the Wagtail admin.

git security advisory reports:

Git uses external "credential helper" programs to store and retrieve passwords or other credentials from secure storage provided by the operating system. Specially-crafted URLs that contain an encoded newline can inject unintended values into the credential helper protocol stream, causing the credential helper to retrieve the password for one server for an HTTP request being made to another server, resulting in credentials for the former being sent to the latter.

NVD reports:

Libntlm through 1.5 relies on a fixed buffer size for tSmbNtlmAuthRequest, tSmbNtlmAuthChallenge, and tSmbNtlmAuthResponse read and write operations, as demonstrated by a stack-based buffer over-read in buildSmbNtlmAuthRequest in smbutil.c for a crafted NTLM request.

Problem Description:

Incomplete packet data validation may result in accessing out-of-bounds memory (CVE-2019-5614) or may access memory after it has been freed (CVE-2019-15874).

Impact:

Access to out of bounds or freed mbuf data can lead to a kernel panic or other unpredictable results.

Twisted developers reports:

All HTTP clients in twisted.web.client now raise a ValueError when called with a method and/or URL that contain invalid characters. This mitigates CVE-2019-12387. Thanks to Alex Brasetvik for reporting this vulnerability.

The HTTP/2 server implementation now enforces TCP flow control on control frame messages and times out clients that send invalid data without reading responses. This closes CVE-2019-9512 (Ping Flood), CVE-2019-9514 (Reset Flood), and CVE-2019-9515 (Settings Flood). Thanks to Jonathan Looney and Piotr Sikora.

twisted.web.http was subject to several request smuggling attacks. Requests with multiple Content-Length headers were allowed (CVE-2020-10108, thanks to Jake Miller from Bishop Fox and ZeddYu Lu for reporting this) and now fail with a 400; requests with a Content-Length header and a Transfer-Encoding header honored the first header (CVE-2020-10109, thanks to Jake Miller from Bishop Fox for reporting this) and now fail with a 400; requests whose Transfer-Encoding header had a value other than "chunked" and "identity" (thanks to ZeddYu Lu) were allowed and now fail with a 400.

The libssh team reports (originally reported by Yasheng Yang from Google):

A malicious client or server could crash the counterpart implemented with libssh AES-CTR ciphers are used and don't get fully initialized. It will crash when it tries to cleanup the AES-CTR ciphers when closing the connection.

The WebKitGTK project reports the following vulnerability.

Processing maliciously crafted web content may lead to arbitrary code execution or application crash (denial of service). Description: A memory corruption issue (use-after-free) was addressed with improved memory handling.

Borja Tarraso reports:

A flaw was found in Ansible 2.7.17 and prior, 2.8.9 and prior, and 2.9.6 and prior when using the Extract-Zip function from the win_unzip module as the extracted file(s) are not checked if they belong to the destination folder. An attacker could take advantage of this flaw by crafting an archive anywhere in the file system, using a path traversal. This issue is fixed in 2.10.

Borja Tarraso reports:

A flaw was found in Ansible 2.7.16 and prior, 2.8.8 and prior, and 2.9.5 and prior when a password is set with the argument "password" of svn module, it is used on svn command line, disclosing to other users within the same node. An attacker could take advantage by reading the cmdline file from that particular PID on the procfs.

Borja Tarraso reports:

A flaw was found in Ansible Engine when using Ansible Vault for editing encrypted files. When a user executes "ansible-vault edit", another user on the same computer can read the old and new secret, as it is created in a temporary file with mkstemp and the returned file descriptor is closed and the method write_data is called to write the existing secret in the file. This method will delete the file before recreating it insecurely. All versions in 2.7.x, 2.8.x and 2.9.x branches are believed to be vulnerable.

Drupal Security Team reports:

The Drupal project uses the third-party library CKEditor, which has released a security improvement that is needed to protect some Drupal configurations.

Vulnerabilities are possible if Drupal is configured to use the WYSIWYG CKEditor for your site's users. An attacker that can createor edit content may be able to exploit this Cross Site Scripting (XSS) vulnerability to target users with access to the WYSIWYG CKEditor, and this may include site admins with privileged access.

The latest versions of Drupal update CKEditor to 4.14 to mitigate the vulnerabilities.

Google Chrome Releases reports:

[1067851] Critical CVE-2020-6457: Use after free in speech recognizer. Reported by Leecraso and Guang Gong of Alpha Lab, Qihoo 360 on 2020-04-04

Lev Stipakov and Gert Doering report:

There is a time frame between allocating peer-id and initializing data channel key (which is performed on receiving push request or on async push-reply) in which the existing peer-id float checks do not work right.

If a "rogue" data channel packet arrives during that time frame from another address and with same peer-id, this would cause client to float to that new address.

The net effect of this behaviour is that the VPN session for the "victim client" is broken. Since the "attacker client" does not have suitable keys, it can not inject or steal VPN traffic from the other session. The time window is small and it can not be used to attack a specific client's session, unless some other way is found to make it disconnect and reconnect first.

Gitlab reports:

NuGet Package and File Disclosure through GitLab Workhorse

Job Artifact Uploads and File Disclosure through GitLab Workhorse

Incorrect membership following group removal

Logging of Praefect tokens

Update Rack dependency

Update OpenSSL dependency

Manuel Pégourié-Gonnard reports:

An attacker with access to precise enough timing and memory access information (typically an untrusted operating system attacking a secure enclave such as SGX or the TrustZone secure world) can fully recover an ECDSA private key after observing a number of signature operations.

RedHat reports:

ceph: secure mode of msgr2 breaks both confidentiality and integrity aspects for long-lived sessions.

ceph: header-splitting in RGW GetObject has a possible XSS.

Jon Siwek of Corelight reports:

This release fixes the following security issue:

• An attacker can crash Zeek remotely via crafted packet sequence.

Google Chrome Releases reports:

This updates includes 32 security fixes, including:

• [1019161] High CVE-2020-6454: Use after free in extensions. Reported by Leecraso and Guang Gong of Alpha Lab, Qihoo 360 on 2019-10-29
• [1043446] High CVE-2020-6423: Use after free in audio. Reported by Anonymous on 2020-01-18
• [1059669] High CVE-2020-6455: Out of bounds read in WebSQL. Reported by Nan Wang(@eternalsakura13) and Guang Gong of Alpha Lab, Qihoo 360 on 2020-03-09
• [1031479] Medium CVE-2020-6430: Type Confusion in V8. Reported by Avihay Cohen @ SeraphicAlgorithms on 2019-12-06
• [1040755] Medium CVE-2020-6456: Insufficient validation of untrusted input in clipboard. Reported by Micha? Bentkowski of Securitum on 2020-01-10
• [852645] Medium CVE-2020-6431: Insufficient policy enforcement in full screen. Reported by Luan Herrera (@lbherrera_) on 2018-06-14
• [965611] Medium CVE-2020-6432: Insufficient policy enforcement in navigations. Reported by David Erceg on 2019-05-21
• [1043965] Medium CVE-2020-6433: Insufficient policy enforcement in extensions. Reported by David Erceg on 2020-01-21
• [1048555] Medium CVE-2020-6434: Use after free in devtools. Reported by HyungSeok Han (DaramG) of Theori on 2020-02-04
• [1032158] Medium CVE-2020-6435: Insufficient policy enforcement in extensions. Reported by Sergei Glazunov of Google Project Zero on 2019-12-09
• [1034519] Medium CVE-2020-6436: Use after free in window management. Reported by Igor Bukanov from Vivaldi on 2019-12-16
• [639173] Low CVE-2020-6437: Inappropriate implementation in WebView. Reported by Jann Horn on 2016-08-19
• [714617] Low CVE-2020-6438: Insufficient policy enforcement in extensions. Reported by Ng Yik Phang on 2017-04-24
• [868145] Low CVE-2020-6439: Insufficient policy enforcement in navigations. Reported by remkoboonstra on 2018-07-26
• [894477] Low CVE-2020-6440: Inappropriate implementation in extensions. Reported by David Erceg on 2018-10-11
• [959571] Low CVE-2020-6441: Insufficient policy enforcement in omnibox. Reported by David Erceg on 2019-05-04
• [1013906] Low CVE-2020-6442: Inappropriate implementation in cache. Reported by B@rMey on 2019-10-12
• [1040080] Low CVE-2020-6443: Insufficient data validation in developer tools. Reported by @lovasoa (Ophir LOJKINE) on 2020-01-08
• [922882] Low CVE-2020-6444: Uninitialized Use in WebRTC. Reported by mlfbrown on 2019-01-17
• [933171] Low CVE-2020-6445: Insufficient policy enforcement in trusted types. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2019-02-18
• [933172] Low CVE-2020-6446: Insufficient policy enforcement in trusted types. Reported by Jun Kokatsu, Microsoft Browser Vulnerability Research on 2019-02-18
• [991217] Low CVE-2020-6447: Inappropriate implementation in developer tools. Reported by David Erceg on 2019-08-06
• [1037872] Low CVE-2020-6448: Use after free in V8. Reported by Guang Gong of Alpha Lab, Qihoo 360 on 2019-12-26

The Squid developers reports:

Improper Input Validation issues in HTTP Request processing (CVE-2020-8449, CVE-2020-8450).

Information Disclosure issue in FTP Gateway (CVE-2019-12528).

Buffer Overflow issue in ext_lm_group_acl helper (CVE-2020-8517).

When parsing certain JSON documents, the json gem (including the one bundled with Ruby) can be coerced into creating arbitrary objects in the target system.

This is the same issue as CVE-2013-0269. The previous fix was incomplete, which addressed JSON.parse(user_input), but didn?t address some other styles of JSON parsing including JSON(user_input) and JSON.parse(user_input, nil).

See CVE-2013-0269 in detail. Note that the issue was exploitable to cause a Denial of Service by creating many garbage-uncollectable Symbol objects, but this kind of attack is no longer valid because Symbol objects are now garbage-collectable. However, creating arbitrary bjects may cause severe security consequences depending upon the application code.

Please update the json gem to version 2.3.0 or later. You can use gem update json to update it. If you are using bundler, please add gem "json", ">= 2.3.0" to your Gemfile.

The HAproxy Project reports:

The main driver for this release is that it contains a fix for a serious vulnerability that was responsibly reported last week by Felix Wilhelm from Google Project Zero, affecting the HPACK decoder used for HTTP/2. CVE-2020-11100 was assigned to this issue.

Google Chrome Releases reports:

This update contains 8 security fixes.

• [1062247] High CVE-2020-6450: Use after free in WebAudio. Reported by Man Yue Mo of Semmle Security Research Team on 2020-03-17
• [1061018] High CVE-2020-6451: Use after free in WebAudio. Reported by Man Yue Mo of Semmle Security Research Team on 2020-03-12
• [1059764] High CVE-2020-6452: Heap buffer overflow in media Reported by asnine on 2020-03-09
• [1066247] Various fixes from internal audits, fuzzing and other initiatives.

Apache Team reports:

SECURITY: CVE-2020-1934

mod_proxy_ftp: Use of uninitialized value with malicious backend FTP server.

SECURITY: CVE-2020-1927

rewrite, core: Set PCRE_DOTALL flag by default to avoid unpredictable matches and substitutions with encoded line break characters. The fix for CVE-2019-10098 was not effective.

The Cacti developers reports:

When guest users have access to realtime graphs, remote code could be executed (CVE-2020-8813).

Lack of escaping on some pages can lead to XSS exposure (CVE-2020-7106).

Remote Code Execution due to input validation failure in Performance Boost Debug Log (CVE-2020-7237).

The GnuTLS project reports:

It was found that GnuTLS 3.6.3 introduced a regression in the DTLS protocol implementation. This caused the DTLS client to not contribute any randomness to the DTLS negotiation breaking the security guarantees of the DTLS protocol.

The PostgreSQL project reports:

Versions Affected: 9.6 - 12

The ALTER ... DEPENDS ON EXTENSION sub-commands do not perform authorization checks, which can allow an unprivileged user to drop any function, procedure, materialized view, index, or trigger under certain conditions. This attack is possible if an administrator has installed an extension and an unprivileged user can CREATE, or an extension owner either executes DROP EXTENSION predictably or can be convinced to execute DROP EXTENSION.

Mediawiki reports:

Security fixes: T246602:jquery.makeCollapsible allows applying event handler to any CSS selector.

Gitlab reports:

Arbitrary File Read when Moving an Issue

Path Traversal in NPM Package Registry

SSRF on Project Import

External Users Can Create Personal Snippet

Triggers Decription Can be Updated by Other Maintainers in Project

Information Disclosure on Confidential Issues Moved to Private Programs

Potential DoS in Repository Archive Download

Blocked Users Can Still Pull/Push Docker Images

Repository Mirroring not Disabled when Feature not Activated

Vulnerability Feedback Page Was Leaking Information on Vulnerabilities

Stored XSS Vulnerability in Admin Feature

Upload Feature Allowed a User to Read Unauthorized Exported Files

Unauthorized Users Are Able to See CI Metrics

Last Pipeline Status of a Merge Request Leaked

Blind SSRF on FogBugz

Update Nokogiri dependency

Jenkins Security Advisory:

Description

(High) SECURITY-1774 / CVE-2020-2160

CSRF protection for any URL could be bypassed

(Medium) SECURITY-1781 / CVE-2020-2161

Stored XSS vulnerability in label expression validation

(Medium) SECURITY-1793 / CVE-2020-2162

Stored XSS vulnerability in file parameters

(Medium) SECURITY-1796 / CVE-2020-2163

Stored XSS vulnerability in list view column headers

phpMyAdmin Team reports:

PMASA-2020-2 SQL injection vulnerability in the user accounts page, particularly when changing a password

PMASA-2020-3 SQL injection vulnerability relating to the search feature

PMASA-2020-4 SQL injection and XSS having to do with displaying results

Removing of the "options" field for the external transformation

Puppetlabs reports:

Puppet Server and PuppetDB provide useful performance and debugging information via their metrics API endpoints. For PuppetDB this may contain things like hostnames. Puppet Server reports resource names and titles for defined types (which may contain sensitive information) as well as function names and class names. Previously, these endpoints were open to the local network.

PE 2018.1.13 & 2019.4.0, Puppet Server 6.9.1 & 5.3.12, and PuppetDB 6.9.1 & 5.2.13 disable trapperkeeper-metrics /v1 metrics API and only allows /v2 access on localhost by default.

Puppetlabs reports:

Previously, Puppet operated on a model that a node with a valid certificate was entitled to all information in the system and that a compromised certificate allowed access to everything in the infrastructure. When a node's catalog falls back to the `default` node, the catalog can be retrieved for a different node by modifying facts for the Puppet run. This issue can be mitigated by setting `strict_hostname_checking = true` in `puppet.conf` on your Puppet master.

Puppet 6.13.0 changes the default behavior for strict_hostname_checking from false to true. It is recommended that Puppet Open Source and Puppet Enterprise users that are not upgrading still set strict_hostname_checking to true to ensure secure behavior.

Problem Description:

Incorrect use of a potentially user-controlled pointer in the kernel allowed vnet jailed users to panic the system and potentially execute aribitrary code in the kernel.

Impact:

Users with root level access (or the PRIV_NET_IFCREATE privilege) can panic the system, or potentially escape the jail or execute arbitrary code with kernel priviliges.

Problem Description:

When a TCP server transmits or retransmits a TCP SYN-ACK segment over IPv6, the Traffic Class field is not initialized. This also applies to challenge ACK segments, which are sent in response to received RST segments during the TCP connection setup phase.

Impact:

For each TCP SYN-ACK (or challenge TCP-ACK) segment sent over IPv6, one byte of kernel memory is transmitted over the network.

Problem Description:

The driver-specific ioctl(2) command handlers in oce(4) failed to check whether the caller has sufficient privileges to perform the corresponding operation.

Impact:

The oce(4) handler permits unprivileged users to send passthrough commands to device firmware.

Problem Description:

A missing NUL-termination check for the jail_set(2) configration option "osrelease" may return more bytes when reading the jail configuration back with jail_get(2) than were originally set.

Impact:

For jails with a non-default setting of children.max > 0 ("nested jails") a superuser inside a jail can create a jail and may be able to read and take advantage of exposed kernel memory.

Problem Description:

The driver-specific ioctl(2) command handlers in ixl(4) failed to check whether the caller has sufficient privileges to perform the corresponding operation.

Impact:

The ixl(4) handler permits unprivileged users to trigger updates to the device's non-volatile memory (NVM).

* ``bleach.clean`` behavior parsing embedded MathML and SVG content with RCDATA tags did not match browser behavior and could result in a mutation XSS.

Calls to ``bleach.clean`` with ``strip=False`` and ``math`` or ``svg`` tags and one or more of the RCDATA tags ``script``, ``noscript``, ``style``, ``noframes``, ``iframe``, ``noembed``, or ``xmp`` in the allowed tags whitelist were vulnerable to a mutation XSS.

* ``bleach.clean`` behavior parsing ``noscript`` tags did not match browser behavior.

Calls to ``bleach.clean`` allowing ``noscript`` and one or more of the raw text tags (``title``, ``textarea``, ``script``, ``style``, ``noembed``, ``noframes``, ``iframe``, and ``xmp``) were vulnerable to a mutation XSS.

Jon Siwek of Corelight reports:

This release addresses the following security issues:

• Potential Denial of Service due to memory leak in DNS TSIG message parsing.
• Potential Denial of Service due to memory leak (or assertion when compiling with assertions enabled) when receiving a second SSH KEX message after a first.
• Potential Denial of Service due to buffer read overflow and/or memory leaks in Kerberos analyzer. The buffer read overflow could occur when the Kerberos message indicates it contains an IPv6 address, but does not send enough data to parse out a full IPv6 address. A memory leak could occur when processing KRB_KDC_REQ KRB_KDC_REP messages for message types that do not match a known/expected type.
• Potential Denial of Service when sending many zero-length SSL/TLS certificate data. Such messages underwent the full Zeek file analysis treatment which is expensive (and meaninguless here) compared to how cheaply one can "create" or otherwise indicate many zero-length contained in an SSL message.
• Potential Denial of Service due to buffer read overflow in SMB transaction data string handling. The length of strings being parsed from SMB messages was trusted to be whatever the message claimed instead of the actual length of data found in the message.
• Potential Denial of Service due to null pointer dereference in FTP ADAT Base64 decoding.
• Potential Denial of Service due buffer read overflow in FTP analyzer word/whitespace handling. This typically won't be a problem in most default deployments of Zeek since the FTP analyzer receives data from a ContentLine (NVT) support analyzer which first null-terminates the buffer used for further FTP parsing.

Albert Astals Cid:

Okular can be tricked into executing local binaries via specially crafted PDF files.

This binary execution can require almost no user interaction.

No parameters can be passed to those local binaries.

We have not been able to identify any binary that will cause actual damage, be it in the hardware or software level, when run without parameters.

We remain relatively confident that for this issue to do any actual damage, it has to run a binary specially crafted. That binary must have been deployed to the user system via another method, be it the user downloading it directly as an email attachment, webpage download, etc. or by the system being already compromised.

MITRE CVE reports:

Django 1.11 before 1.11.29, 2.2 before 2.2.11, and 3.0 before 3.0.4 allows SQL Injection if untrusted data is used as a tolerance parameter in GIS functions and aggregates on Oracle. By passing a suitably crafted tolerance to GIS functions and aggregates on Oracle, it was possible to break escaping and inject malicious SQL.

Gitlab reports:

Email Confirmation not Required on Sign-up

Matrix developers report:

[The 1.11.1] release includes a security fix impacting installations using Single Sign-On (i.e. SAML2 or CAS) for authentication. Administrators of such installations are encouraged to upgrade as soon as possible.

Node.js reports:

Updates are now available for all active Node.js release lines for the following issues.

HTTP request smuggling using malformed Transfer-Encoding header (Critical) (CVE-2019-15605)HTTP request smuggling using malformed Transfer-Encoding header (Critical) (CVE-2019-15605)

Affected Node.js versions can be exploited to perform HTTP desync attacks and deliver malicious payloads to unsuspecting users. The payloads can be crafted by an attacker to hijack user sessions, poison cookies, perform clickjacking, and a multitude of other attacks depending on the architecture of the underlying system.

HTTP header values do not have trailing OWS trimmed (High) (CVE-2019-15606)

Optional whitespace should be trimmed from HTTP header values. Its presence may allow attackers to bypass security checks based on HTTP header values.

Remotely trigger an assertion on a TLS server with a malformed certificate string (High) (CVE-2019-15604)

Connecting to a NodeJS TLS server with a client certificate that has a type 19 string in its subjectAltName will crash the TLS server if it tries to read the peer certificate.

Strict HTTP header parsing (None)

Increase the strictness of HTTP header parsing. There are no known vulnerabilities addressed, but lax HTTP parsing has historically been a source of problems. Some commonly used sites are known to generate invalid HTTP headers, a --insecure-http-parser CLI option or insecureHTTPParser http option can be used if necessary for interoperability, but is not recommended.

SaltStack reports:

With the Salt NetAPI enabled in addition to having a SSH roster defined, unauthenticated access is possible when specifying the client as SSH.

Additionally, when the raw_shell option is specified any arbitrary command may be run on the Salt master when specifying SSH options.

The Gitea Team reports for release 1.11.0:

• Never allow an empty password to validate (#9682) (#9683)
• Prevent redirect to Host (#9678) (#9679)
• Swagger hide search field (#9554)
• Add "search" to reserved usernames (#9063)
• Switch to fomantic-ui (#9374)
• Only serve attachments when linked to issue/release and if accessible by user (#9340)

The Gitea Team reports for release 1.11.2:

• Ensure only own addresses are updated (#10397) (#10399)
• Logout POST action (#10582) (#10585)
• Org action fixes and form cleanup (#10512) (#10514)
• Change action GETs to POST (#10462) (#10464)
• Fix admin notices (#10480) (#10483)
• Change admin dashboard to POST (#10465) (#10466)
• Update markbates/goth (#10444) (#10445)
• Update crypto vendors (#10385) (#10398)

Gitlab reports:

Directory Traversal to Arbitrary File Read

Account Takeover Through Expired Link

Server Side Request Forgery Through Deprecated Service

Group Two-Factor Authentication Requirement Bypass

Stored XSS in Merge Request Pages

Stored XSS in Merge Request Submission Form

Stored XSS in File View

Stored XSS in Grafana Integration

Contribution Analytics Exposed to Non-members

Incorrect Access Control in Docker Registry via Deploy Tokens

Denial of Service via Permission Checks

Denial of Service in Design For Public Issue

Incorrect Access Control via LFS Import

Unescaped HTML in Header

Private Merge Request Titles Leaked via Widget

Project Namespace Exposed via Vulnerability Feedback Endpoint

Denial of Service Through Recursive Requests

Project Authorization Not Being Updated

Incorrect Permission Level For Group Invites

Disclosure of Private Group Epic Information

User IP Address Exposed via Badge images

nwtime.org reports:

Three ntp vulnerabilities, Depending on configuration, may have little impact up to termination of the ntpd process.

NTP Bug 3610: Process_control() should exit earlier on short packets. On systems that override the default and enable ntpdc (mode 7) fuzz testing detected that a short packet will cause ntpd to read uninitialized data.

NTP Bug 3596: An unauthenticated unmonitored ntpd is vulnerable to attack on IPv4 with highly predictable transmit timestamps. An off-path attacker who can query time from the victim's ntp which receives time from an unauthenticated time source must be able to send from a spoofed IPv4 address of upstream ntp server and and the victim must be able to process a large number of packets with the spoofed IPv4 address of the upstream server. After eight or more successful attacks in a row the attacker can either modify the victim's clock by a small amount or cause ntpd to terminate. The attack is especially effective when unusually short poll intervals have been configured.

NTP Bug 3592: The fix for https://bugs.ntp.org/3445 introduced a bug such that a ntp can be prevented from initiating a time volley to its peer resulting in a DoS.

All three NTP bugs may result in DoS or terimation of the ntp daemon.

Librsvg2 developers reports:

Backport the following fixes from 2.46.x:

Librsvg now has limits on the number of loaded XML elements, and the number of referenced elements within an SVG document. This is to mitigate malicious SVGs which try to consume all memory, and those which try to consume an exponential amount of CPU time.

Fix stack exhaustion with circular references in elements.

Fix a denial-of-service condition from exponential explosion of rendered elements, through nested use of SVG elements in malicious SVGs. This is similar to the XML "billion laughs attack" but for SVG instancing.

qflb.wu of DBAPPSecurity reports:

Ihe insert_note_steps function in readmidi.c in TiMidity++ 2.14.0 can cause a denial of service(divide-by-zero error and application crash) via a crafted mid file.

The resample_gauss function in resample.c in TiMidity++ 2.14.0 can cause a denial of service(heap-buffer-overflow) via a crafted mid file.

The play_midi function in playmidi.c in TiMidity++ 2.14.0 can cause a denial of service(large loop and CPU consumption) via a crafted mid file.

Community reports:

8.1.1 and 8.2.0 users check ENABLE_REMOTE_JMX_OPTS setting

Apache Solr RCE vulnerability due to bad config default

Apache Solr RCE through VelocityResponseWriter

OpenSMTPD developers reports:

An out of bounds read in smtpd allows an attacker to inject arbitrary commands into the envelope file which are then executed as root. Separately, missing privilege revocation in smtpctl allows arbitrary commands to be run with the _smtpq group.

An unprivileged local attacker can read the first line of an arbitrary file (for example, root's password hash in /etc/master.passwd) or the entire contents of another user's file (if this file and /var/spool/smtpd/ are on the same filesystem).

Janos Follath reports:

If Mbed TLS is running in an SGX enclave and the adversary has control of the main operating system, they can launch a side channel attack to recover the RSA private key when it is being imported.

The attack only requires access to fine grained measurements to cache usage. Therefore the attack might be applicable to a scenario where Mbed TLS is running in TrustZone secure world and the attacker controls the normal world or possibly when Mbed TLS is part of a hypervisor and the adversary has full control of a guest OS.

Qualys reports:

.

Qualys reports:

We discovered a minor vulnerability in OpenSMTPD, OpenBSD's mail server: an unprivileged local attacker can read the first line of an arbitrary file (for example, root's password hash in /etc/master.passwd) or the entire contents of another user's file (if this file and /var/spool/smtpd/ are on the same filesystem).

Janos Follath reports:

Our bignum implementation is not constant time/constant trace, so side channel attacks can retrieve the blinded value, factor it (as it is smaller than RSA keys and not guaranteed to have only large prime factors), and then, by brute force, recover the key.

The WeeChat project reports:

Buffer overflow when receiving a malformed IRC message 324 (channel mode). (CVE-2020-8955)

Buffer overflow when a new IRC message 005 is received with longer nick prefixes.

Crash when receiving a malformed IRC message 352 (WHO).

The WebKitGTK project reports multiple vulnerabilities.

Gitlab reports:

Incorrect membership handling of group sharing feature

Aki Tuomi reports:

lib-smtp doesn't handle truncated command parameters properly, resulting in infinite loop taking 100% CPU for the process. This happens for LMTP (where it doesn't matter so much) and also for submission-login where unauthenticated users can trigger it.

Aki also reports:

Snippet generation crashes if: message is large enough that message-parser returns multiple body blocks The first block(s) don't contain the full snippet (e.g. full of whitespace) input ends with '>'

Reno Robert reports:

FreeBSD uses a two-process model for running a VM. For booting non-FreeBSD guests, a modified grub-emu is used (grub-bhyve). Grub-bhyve executes command from guest grub.cfg file. This is a security problem because grub was never written to handle inputs from OS as untrusted. In the current design, grub and guest OS works across trust boundaries. This exposes a grub to untrusted inputs from guest.

grub-bhyve (emu) is built without SDL graphics support which reduces lot of gfx attack surface, however font loading code is still accessible. Guest can provide arbitrary font file, which is parsed by grub-bhyve running as root.

In grub-core/font/font.c, read_section_as_string() allocates section->length + 1 bytes of memory. However, untrusted section->length is an unsigned 32-bit number, and the result can overflow to malloc(0). This can result in a controlled buffer overflow via the 'loadfont' command in a guest VM grub2.cfg, eventually leading to privilege escalation from guest to host.

Reno Robert also reports:

GRUB supports commands to read and write addresses of choice. In grub-bhyve, these commands provide a way to write to arbitrary virtual addresses within the grub-bhyve process. This is another way for a guest grub2.cfg, run by the host, to eventually escalate privileges.

These vulnerabilities are mitigated by disabling the 'loadfont', 'write_dword', 'read_dword', 'inl', 'outl', and other width variants of the same functionality in grub2-bhyve.

There is also work in progress to sandbox the grub-bhyve utility such that an escaped guest ends up with nobody:nobody in a Capsium sandbox. It is not included in 0.40_8.

Mitre reports:

In libexif, there is a possible out of bounds write due to an integer overflow. This could lead to remote escalation of privilege in the media content provider with no additional execution privileges needed. User interaction is needed for exploitation.

Adobe reports:

• This update resolves a type confusion vulnerability that could lead to arbitrary code execution (CVE-2020-3757).

NGINX Team reports:

NGINX before 1.17.7, with certain error_page configurations, allows HTTP request smuggling, as demonstrated by the ability of an attacker to read unauthorized web pages in environments where NGINX is being fronted by a load balancer.

Upstream ksh93 maintainer Siteshwar Vashisht reports:

A flaw was found in the way ksh evaluates certain environment variables. An attacker could use this flaw to override or bypass environment restrictions to execute shell commands. Services and applications that allow remote unauthenticated attackers to provide one of those environment variables could allow them to exploit this issue remotely.

Micah Snyder reports:

A denial-of-service (DoS) condition may occur when using the optional credit card data-loss-prevention (DLP) feature. Improper bounds checking of an unsigned variable resulted in an out-of-bounds read, which causes a crash.

MITRE CVE reports:

Django 1.11 before 1.11.28, 2.2 before 2.2.10, and 3.0 before 3.0.3 allows SQL Injection if untrusted data is used as a StringAgg delimiter (e.g., in Django applications that offer downloads of data as a series of rows with a user-specified column delimiter). By passing a suitably crafted delimiter to a contrib.postgres.aggregates.StringAgg instance, it was possible to break escaping and inject malicious SQL.

The libssh team reports:

In an environment where a user is only allowed to copy files and not to execute applications, it would be possible to pass a location which contains commands to be executed in additon.

When the libssh SCP client connects to a server, the scp command, which includes a user-provided path, is executed on the server-side. In case the library is used in a way where users can influence the third parameter of ssh_scp_new(), it would become possible for an attacker to inject arbitrary commands, leading to a compromise of the remote target.

Oracle reports:

This Critical Patch Update contains 17 new security fixes for Oracle MySQL. 5 of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials.

MariaDB reports:

Difficult to exploit vulnerability allows unauthenticated attacker with network access via multiple protocols to compromise MySQL Client.

Gitlab reports:

Path Traversal to Arbitrary File Read

User Permissions Not Validated in ProjectExportWorker

XSS Vulnerability in File API

Package and File Disclosure through GitLab Workhorse

XSS Vulnerability in Create Groups

Issue and Merge Request Activity Counts Exposed

Email Confirmation Bypass Using AP

Disclosure of Forked Private Project Source Code

Private Project Names Exposed in GraphQL queries

Disclosure of Issues and Merge Requests via Todos

Denial of Service via AsciiDoc

Last Pipeline Status Exposed

Arbitrary Change of Pipeline Status

Grafana Token Displayed in Plaintext

Update excon gem

Update rdoc gem

Update rack-cors gem

Update rubyzip gem

The Apache SpamAssassin project reports:

A nefarious rule configuration (.cf) files can be configured to run system commands. This issue is less stealthy and attempts to exploit the issue will throw warnings.

Thanks to Damian Lukowski at credativ for reporting the issue ethically. With this bug unpatched, exploits can be injected in a number of scenarios though doing so remotely is difficult. In addition to upgrading to SA 3.4.4, we again recommend that users should only use update channels or 3rd party .cf files from trusted places.

Todd C. Miller reports:

Sudo's pwfeedback option can be used to provide visual feedback when the user is inputting their password. For each key press, an asterisk is printed. This option was added in response to user confusion over how the standard Password: prompt disables the echoing of key presses. While pwfeedback is not enabled by default in the upstream version of sudo, some systems, such as Linux Mint and Elementary OS, do enable it in their default sudoers files.

Due to a bug, when the pwfeedback option is enabled in the sudoers file, a user may be able to trigger a stack-based buffer overflow. This bug can be triggered even by users not listed in the sudoers file. There is no impact unless pwfeedback has been enabled.

the Apache Spamassassin project reports:

nefarious rule configuration (.cf) files can be configured to run system commands similar to CVE-2018-11805. This issue is less stealthy and attempts to exploit the issue will throw warnings.

OpenSMTPD developers report:

An incorrect check allows an attacker to trick mbox delivery into executing arbitrary commands as root and lmtp delivery into executing arbitrary commands as an unprivileged user

Problem Description:

A programming error allows an attacker who can specify a URL with a username and/or password components to overflow libfetch(3) buffers.

Impact:

An attacker in control of the URL to be fetched (possibly via HTTP redirect) may cause a heap buffer overflow, resulting in program misbehavior or malicious code execution.

A programming error allows an attacker who can specify a URL with a username and/or password components to overflow libfetch(3) buffers.

Problem Description:

A missing check means that an attacker can reinject an old packet and it will be accepted and processed by the IPsec endpoint.

Impact:

The impact depends on the higher-level protocols in use over IPsec. For example, an attacker who can capture and inject packets could cause an action that was intentionally performed once to be repeated.

Problem Description:

Due to incorrect initialization of a stack data structure, up to 20 bytes of kernel data stored previously stored on the stack will be exposed to a crashing user process.

Impact:

Sensitive kernel data may be disclosed.

Jenkins Security Advisory:

Description

(High) SECURITY-1682 / CVE-2020-2099

Inbound TCP Agent Protocol/3 authentication bypass

(Medium) SECURITY-1641 / CVE-2020-2100

Jenkins vulnerable to UDP amplification reflection attack

(Medium) SECURITY-1659 / CVE-2020-2101

Non-constant time comparison of inbound TCP agent connection secret

(Medium) SECURITY-1660 / CVE-2020-2102

Non-constant time HMAC comparison

(Medium) SECURITY-1695 / CVE-2020-2103

Diagnostic page exposed session cookies

(Medium) SECURITY-1650 / CVE-2020-2104

Memory usage graphs accessible to anyone with Overall/Read

(Low) SECURITY-1704 / CVE-2020-2105

Jenkins REST APIs vulnerable to clickjacking

(Medium) SECURITY-1680 / CVE-2020-2106

Stored XSS vulnerability in Code Coverage API Plugin

(Medium) SECURITY-1565 / CVE-2020-2107

Fortify Plugin stored credentials in plain text

(High) SECURITY-1719 / CVE-2020-2108

XXE vulnerability in WebSphere Deployer Plugin

The Samba Team reports:

CVE-2019-14902

The implementation of ACL inheritance in the Samba AD DC was not complete, and so absent a 'full-sync' replication, ACLs could get out of sync between domain controllers.

CVE-2019-14907

When processing untrusted string input Samba can read past the end of the allocated buffer when printing a "Conversion error" message to the logs.

CVE-2019-19344

During DNS zone scavenging (of expired dynamic entries) there is a read of memory after it has been freed.

The WebKitGTK project reports multiple vulnerabilities.

Pillow developers report:

This release addresses several security problems, as well as addressing CVE-2019-19911.

CVE-2019-19911 is regarding FPX images. If an image reports that it has a large number of bands, a large amount of resources will be used when trying to process the image. This is fixed by limiting the number of bands to those usable by Pillow.

Buffer overruns were found when processing an SGI, PCX or FLI image. Checks have been added to prevent this.

Overflow checks have been added when calculating the size of a memory block to be reallocated in the processing of a TIFF image.

The Gitea Team reports:

• Hide credentials when submitting migration
• Never allow an empty password to validate
• Prevent redirect to Host
• Hide public repos owned by private orgs

Intel reports:

.A potential security vulnerability in Intel(R) Processor Graphics may allow information disclosure. Intel is releasing software updates to mitigate this potential vulnerability.

Description: Insufficient control flow in certain data structures for some Intel(R) Processors with Intel(R) Processor Graphics may allow an unauthenticated user to potentially enable information disclosure via local access.

This patch provides mitigation for Gen9 hardware only. Patches for Gen7 and Gen7.5 will be provided later. Note that Gen8 is not impacted due to a previously implemented workaround. The mitigation involves using an existing hardware feature to forcibly clear down all EU state at each context switch.

Art Manion and Will Dormann report:

By using an older and less-secure form of open(), it is possible for untrusted template files to cause reads/writes outside of the template directories. This vulnerability is a component of the recent Citrix exploit.

Gitlab reports:

Private objects exposed through project importi

The phpMyAdmin development team reports:

A SQL injection flaw has been discovered in the user accounts page

Lilith of Cisco Talos reports:

A code execution vulnerability exists in the directory rehashing functionality of E2fsprogs e2fsck 1.45.4. A specially crafted ext4 directory can cause an out-of-bounds write on the stack, resulting in code execution. An attacker can corrupt a partition to trigger this vulnerability.

Theodore Y. Ts'o reports:

E2fsprogs 1.45.5 [...:] Fix a potential out of bounds write when checking a maliciously corrupted file system. This is probably not exploitable on 64-bit platforms, but may be exploitable on 32-bit binaries depending on how the compiler lays out the stack variables. (Addresses CVE-2019-5188)

The cacti developers reports:

When viewing graphs, some input variables are not properly checked (SQL injection possible).

Multiple instances of lib/functions.php are affected by unsafe deserialization of user-controlled data to populate arrays. An authenticated attacker could use this to influence object data values and control actions taken by Cacti or potentially cause memory corruption in the PHP module.

SO-AND-SO reports:

Group Maintainers Can Update/Delete Group Runners Using API

GraphQL Queries Can Hang the Application

Unauthorized Users Have Access to Milestones of Releases

Private Group Name Revealed Through Protected Tags API

Users Can Publish Reviews on Locked Merge Requests

DoS in the Issue and Commit Comments Pages

Project Name Disclosed Through Unsubscribe Link

Private Project Name Disclosed Through Notification Settings

National Vulnerability Database:

There's a possible information leak / session hijack vulnerability in Rack (RubyGem rack). This vulnerability is patched in versions 1.6.12 and 2.0.8. Attackers may be able to find and hijack sessions by using timing attacks targeting the session id. Session ids are usually stored and indexed in a database that uses some kind of scheme for speeding up lookups of that session id. By carefully measuring the amount of time it takes to look up a session, an attacker may be able to find a valid session id and hijack the session. The session id itself may be generated randomly, but the way the session is indexed by the backing store does not use a secure comparison.

Cary Phillips reports:

OpenEXR (IlmBase) v2.4.0 fixes the following security vulnerabilities:

• CVE-2018-18444 Issue #351 Out of Memory
• CVE-2018-18443 Issue #350 heap-buffer-overflow

The relevant patches have been backported to the FreeBSD ports.

wordpress developers reports:

Four security issues affect WordPress versions 5.3 and earlier; version 5.3.1 fixes them, so youll want to upgrade. If you havent yet updated to 5.3, there are also updated versions of 5.2 and earlier that fix the security issues. -Props to Daniel Bachhuber for finding an issue where an unprivileged user could make a post sticky via the REST API. -Props to Simon Scannell of RIPS Technologies for finding and disclosing an issue where cross-site scripting (XSS) could be stored in well-crafted links. -Props to the WordPress.org Security Team for hardening wp_kses_bad_protocol() to ensure that it is aware of the named colon attribute. -Props to Nguyen The Duc for discovering a stored XSS vulnerability using block editor content.

Typo3 core team reports:

It has been discovered that the output of field validation errors in the Form Framework is vulnerable to cross-site scripting.

It has been discovered that t3:// URL handling and typolink functionality are vulnerable to cross-site scripting. Not only regular backend forms are affected but also frontend extensions which use the rendering with typolink.

It has been discovered that the output table listing in the Files backend module is vulnerable to cross-site scripting when a file extension contains malicious sequences. Access to the file system of the server - either directly or through synchronization - is required to exploit the vulnerability.

It has been discovered that the extraction of manually uploaded ZIP archives in Extension Manager is vulnerable to directory traversal. Admin privileges are required in order to exploit this vulnerability. Since TYPO3 v9 LTS, System Maintainer privileges are required as well.

Failing to properly escape user submitted content, class QueryGenerator is vulnerable to SQL injection. Having system extension ext:lowlevel installed and a valid backend user having administrator privileges are required to exploit this vulnerability.

It has been discovered that classes QueryGenerator and QueryView are vulnerable to insecure deserialization. Requirements for successfully exploiting this vulnerability (one of the following): - having system extension ext:lowlevel (Backend Module: DB Check) installed and valid backend user having administrator privileges - having system extension ext:sys_action installed and valid backend user having limited privileges

TYPO3 allows to upload files either in the backend user interface as well as in custom developed extensions. To reduce the possibility to upload potential malicious code TYPO3 uses the fileDenyPattern to deny e.g. user submitted PHP scripts from being persisted. Besides that it is possible for any editor to upload file assets using the file module (fileadmin) or changing their avatar image shown in the TYPO3 backend. Per default TYPO3 allows to upload and store HTML and SVG files as well using the mentioned functionalities. Custom extension implementations probably would also accept those files when only the fileDenyPattern is evaluated. Since HTML and SVG files - which might contain executable JavaScript code per W3C standard - could be directly displayed in web clients, the whole web application is exposed to be vulnerable concerning Cross-Site Scripting. Currently the following scenarios are known - given an authenticated regular editor is able to upload files using the TYPO3 backend: - directly target a potential victim to a known public resource in a URL, e.g. /fileadmin/malicious.svg or /fileadmin/malicious.html - using the TypoScript content object ?SVG? (implemented in class ScalableVectorGraphicsContentObject) having renderMode set to inline for SVG files (available since TYPO3 v9.0) - custom implementations that directly output and render markup of HTML and SVG files SVG files that are embedded using an img src=?malicious.svg? tag are not vulnerable since potential scripts are not executed in these scenarios (see https://www.w3.org/wiki/SVG_Security). The icon API of TYPO3 is not scope of this announcement since SVG icons need to be registered using an individual implementation, which is not considered as user submitted content.

It has been discovered that request handling in Extbase can be vulnerable to insecure deserialization. User submitted payload has to be signed with a corresponding HMAC-SHA1 using the sensitive TYPO3 encryptionKey as secret - invalid or unsigned payload is not deserialized. However, since sensitive information could have been leaked by accident (e.g. in repositories or in commonly known and unprotected backup files), there is the possibility that attackers know the private encryptionKey and are able to calculate the required HMAC-SHA1 to allow a malicious payload to be deserialized. Requirements for successfully exploiting this vulnerability (all of the following): - rendering at least one Extbase plugin in the frontend - encryptionKey has been leaked (from LocalConfiguration.php or corresponding .env file).

Drupal Security Team reports:

A visit to install.php can cause cached data to become corrupted. This could cause a site to be impaired until caches are rebuilt.

Drupal 8 core's file_save_upload() function does not strip the leading and trailing dot ('.') from filenames, like Drupal 7 did. Users with the ability to upload files with any extension in conjunction with contributed modules may be able to use this to upload system files such as .htaccess in order to bypass protections afforded by Drupal's default .htaccess file. After this fix, file_save_upload() now trims leading and trailing dots from filenames.

The Media Library module has a security vulnerability whereby it doesn't sufficiently restrict access to media items in certain configurations.

The Drupal project uses the third-party library Archive_Tar, which has released a security-related feature that impacts some Drupal configurations. Multiple vulnerabilities are possible if Drupal is configured to allow .tar, .tar.gz, .bz2 or .tlz file uploads and processes them. The latest versions of Drupal update Archive_Tar to 1.4.9 to mitigate the file processing vulnerabilities.

Ted Y. Ts'o reports:

A maliciously corrupted file systems can trigger buffer overruns in the quota code used by e2fsck.

The OpenSSL project reports:

rsaz_512_sqr overflow bug on x86_64 (CVE-2019-1551) (Low)

There is an overflow bug in the x64_64 Montgomery squaring procedure used in exponentiation with 512-bit moduli. No EC algorithms are affected. Analysis suggests that attacks against 2-prime RSA1024, 3-prime RSA1536, and DSA1024 as a result of this defect would be very difficult to perform and are not believed likely. Attacks against DH512 are considered just feasible. However, for an attack the target would have to re-use the DH512 private key, which is not recommended anyway. Also applications directly using the low level API BN_mod_exp may be affected if they use BN_FLG_CONSTTIME.

Matrix developers report:

The [synapse 1.7.1] release includes several security fixes as well as a fix to a bug exposed by the security fixes. All previous releases of Synapse are affected. Administrators are encouraged to upgrade as soon as possible.

• Fix a bug which could cause room events to be incorrectly authorized using events from a different room.
• Fix a bug causing responses to the /context client endpoint to not use the pruned version of the event.
• Fix a cause of state resets in room versions 2 onwards.

the Apache Spamassassin project reports:

An input validation error of user-supplied input parsing multipart emails. Specially crafted emails can consume all resources on the system.

A local user is able to execute arbitrary shell commands through specially crafted nefarious CF files.

Aki Tuomi reports

Mail with group address as sender will cause a signal 11 crash in push notification drivers. Group address as recipient can cause crash in some drivers.

The Samba Team reports:

CVE-2019-14861:

An authenticated user can crash the DCE/RPC DNS management server by creating records with matching the zone name.

CVE-2019-14870:

The DelegationNotAllowed Kerberos feature restriction was not being applied when processing protocol transition requests (S4U2Self), in the AD DC KDC.

Gitlab reports:

Path traversal with potential remote code execution

Disclosure of private code via Elasticsearch integration

Update Git dependency

Cedric Buissart (Red Hat) reports:

A flaw was found in, ghostscript versions prior to 9.50, in the .pdf_hook_DSC_Creator procedure where it did not properly secure its privileged calls, enabling scripts to bypass `-dSAFER` restrictions. A specially crafted PostScript file could disable security protection and then have access to the file system, or execute arbitrary commands.

A flaw was found in all ghostscript versions 9.x before 9.50, in the .setuserparams2 procedure where it did not properly secure its privileged calls, enabling scripts to bypass `-dSAFER` restrictions. A specially crafted PostScript file could disable security protection and then have access to the file system, or execute arbitrary commands.

A flaw was found in ghostscript, versions 9.x before 9.50, in the setsystemparams procedure where it did not properly secure its privileged calls, enabling scripts to bypass `-dSAFER` restrictions. A specially crafted PostScript file could disable security protection and then have access to the file system, or execute arbitrary commands.

A flaw was found in, ghostscript versions prior to 9.50, in the .pdfexectoken and other procedures where it did not properly secure its privileged calls, enabling scripts to bypass `-dSAFER` restrictions. A specially crafted PostScript file could disable security protection and then have access to the file system, or execute arbitrary commands.

the phpmyadmin team reports:

This security fix is part of an ongoing effort to improve the security of the Designer feature and is designated PMASA-2019-5. There is also an improvement for how we sanitize git version information shown on the home page.

Intel reports:

As part of IPU 2019.2, INTEL-SA-00242 advises that insufficient access control may allow an authenticated user to potentially enable escalation of privilege via local access.

INTEL-SA-00260 advises that insufficient access control may allow an authenticated user to potentially enable denial of service via local access.

Django release reports:

CVE-2019-19118: Privilege escalation in the Django admin.

Since Django 2.1, a Django model admin displaying a parent model with related model inlines, where the user has view-only permissions to a parent model but edit permissions to the inline model, would display a read-only view of the parent model but editable forms for the inline.

Submitting these forms would not allow direct edits to the parent model, but would trigger the parent model's save() method, and cause pre and post-save signal handlers to be invoked. This is a privilege escalation as a user who lacks permission to edit a model should not be able to trigger its save-related signals.

Gitlab reports:

Unauthorized access to grafana metrics

Update Mattermost dependency

Matrix developers report:

Clean up local threepids from user on account deactivation.

Gitlab reports:

Path traversal with potential remote code execution

Private objects exposed through project import

Disclosure of notes via Elasticsearch integration

Disclosure of comments via Elasticsearch integration

DNS Rebind SSRF in various chat notifications

Disclosure of vulnerability status in dependency list

Disclosure of commit count in Cycle Analytics

Exposure of related branch names

Tags pushes from blocked users

Branches and Commits exposed to Guest members via integration

IDOR when adding users to protected environments

Former project members able to access repository information

Unauthorized access to grafana metrics

Todos created for former project members

Update Mattermost dependency

Disclosure of AWS secret keys on certain Admin pages

Stored XSS in Group and User profile fields

Forked project information disclosed via Project API

Denial of Service in the issue and commit comment pages

Tokens stored in plaintext

The WebKitGTK project reports multiple vulnerabilities.

NIST reports: (by search in the range 2018/01/01 - 2019/11/10):

urllib3 before version 1.23 does not remove the Authorization HTTP header when following a cross-origin redirect (i.e., a redirect that differs in host, port, or scheme). This can allow for credentials in the Authorization header to be exposed to unintended hosts or transmitted in cleartext.

In the urllib3 library through 1.24.1 for Python, CRLF injection is possible if the attacker controls the request parameter.

The urllib3 library before 1.24.2 for Python mishandles certain cases where the desired set of CA certificates is different from the OS store of CA certificates, which results in SSL connections succeeding in situations where a verification failure is the correct outcome. This is related to use of the ssl_context, ca_certs, or ca_certs_dir argument.

Micah Snyder reports:

A Denial-of-Service (DoS) vulnerability may occur when scanning a specially crafted email file as a result of excessively long scan times. The issue is resolved by implementing several maximums in parsing MIME messages and by optimizing use of memory allocation.

Intel discovered a previously published erratum on some Intel platforms can be exploited by malicious software to potentially cause a denial of service by triggering a machine check that will crash or hang the system.

Malicious guest operating systems may be able to crash the host.

Starting with version 1.26, the devcpu-data port/package includes updates and mitigations for the following technical and security advisories (depending on CPU model).

Intel TSX Updates (TAA) CVE-2019-11135 Voltage Modulation Vulnerability CVE-2019-11139 MD_CLEAR Operations CVE-2018-12126 CVE-2018-12127 CVE-2018-12130 CVE-2018-11091 TA Indirect Sharing CVE-2017-5715 EGETKEY CVE-2018-12126 CVE-2018-12127 CVE-2018-12130 CVE-2018-11091 JCC SKX102 Erratum

Updated microcode includes mitigations for CPU issues, but may also cause a performance regression due to the JCC erratum mitigation. Please visit http://www.intel.com/benchmarks for further information.

Please visit http://www.intel.com/security for detailed information on these advisories as well as a list of CPUs that are affected.

Operating a CPU without the latest microcode may result in erratic or unpredictable behavior, including system crashes and lock ups. Certain issues listed in this advisory may result in the leakage of privileged system information to unprivileged users. Please refer to the security advisories listed above for detailed information.

Unbound Security Advisories:

Recent versions of Unbound contain a vulnerability that can cause shell code execution after receiving a specially crafted answer. This issue can only be triggered if unbound was compiled with --enable-ipsecmod support, and ipsecmod is enabled and used in the configuration.

The Asterisk project reports:

A remote authenticated Asterisk Manager Interface (AMI) user without system authorization could use a specially crafted Originate AMI request to execute arbitrary system commands.

The Asterisk project reports:

If Asterisk receives a re-invite initiating T.38 faxing and has a port of 0 and no c line in the SDP, a crash will occur.

The Asterisk project reports:

A SIP request can be sent to Asterisk that can change a SIP peers IP address. A REGISTER does not need to occur, and calls can be hijacked as a result. The only thing that needs to be known is the peers name; authentication details such as passwords do not need to be known. This vulnerability is only exploitable when the nat option is set to the default, or auto_force_rport.

The Gitea Team reports:

This release contains five security fixes, so we recommend updating:

• Fix issue with user.fullname
• Ignore mentions for users with no access
• Be more strict with git arguments
• Extract the username and password from the mirror url
• Reserve .well-known username

Squid Team reports:

Problem Description: Due to incorrect data management Squid is vulnerable to a information disclosure when processing HTTP Digest Authentication.

Severity: Nonce tokens contain the raw byte value of a pointer which sits within heap memory allocation. This information reduces ASLR protections and may aid attackers isolating memory areas to target for remote code execution attacks.

Sergey Poznyakoff reports:

This stable release fixes several potential vulnerabilities

CVE-2015-1197: cpio, when using the --no-absolute-filenames option, allows local users to write to arbitrary files via a symlink attack on a file in an archive.

CVE-2016-2037: The cpio_safer_name_suffix function in util.c allows remote attackers to cause a denial of service (out-of-bounds write) via a crafted cpio file.

CVE-2019-14866: Improper input validation when writing tar header fields leads to unexpected tar generation.

National Vulnerability Database:

CVE-2017-8372: The mad_layer_III function in layer3.c in Underbit MAD libmad 0.15.1b, if NDEBUG is omitted, allows remote attackers to cause a denial of service (assertion failure and application exit) via a crafted audio file.

CVE-2017-8373: The mad_layer_III function in layer3.c in Underbit MAD libmad 0.15.1b allows remote attackers to cause a denial of service (heap-based buffer overflow and application crash) or possibly have unspecified other impact via a crafted audio file.

CVE-2017-8374: The mad_bit_skip function in bit.c in Underbit MAD libmad 0.15.1b allows remote attackers to cause a denial of service (heap-based buffer over-read and application crash) via a crafted audio file.

wordpress developers reports:

Props to Evan Ricafort for finding an issue where stored XSS (cross-site scripting) could be added via the Customizer.

rops to J.D. Grimes who found and disclosed a method of viewing unauthenticated posts.

Props to Weston Ruter for finding a way to create a stored XSS to inject Javascript into style tags.

rops to David Newman for highlighting a method to poison the cache of JSON GET requests via the Vary: Origin header.

Props to Eugene Kolodenker who found a server-side request forgery in the way that URLs are validated.

Props to Ben Bidner of the WordPress Security Team who discovered issues related to referrer validation in the admin.

Google Chrome Releases reports:

Four security issues were fixed, including:

• [1021723] Various fixes from internal audits, fuzzing and other initiatives

Sonatype reports:

Several RCE vulnerabilities have been found and corrected in 2.14.15:

CVE-2019-16530: An attacker with elevated privileges can upload a specially crafted file. That file can contain commands that will be executed on the system, with the same privileges as the user running the server.

CVE-2019-15893: A Remote Code Execution vulnerability has been discovered in Nexus Repository Manager requiring immediate action. The vulnerability allows for an attacker with administrative access to NXRM to create repostories that can grant access to read/execute system data outside the scope of NXRM.

CVE-2019-5475: A vulnerability has been found that can allow user's with administrative privileges to run processes on the target server, that the nxrm os user has access to.

The PHP project reports:

The PHP development team announces the immediate availability of PHP 7.3.11. This is a security release which also contains several bug fixes.

The PHP development team announces the immediate availability of PHP 7.2.24. This is a security release which also contains several bug fixes.

The PHP development team announces the immediate availability of PHP 7.1.33. This is a security release which also contains several bug fixes.

Mediawiki reports:

Security fixes: T230402, CVE-2019-16738 SECURITY: Add permission check for suppressed account to Special:Redirect.

mitre reports

cdf_read_property_info in cdf.c in file through 5.37 does not restrict the number of CDF_VECTOR elements, which allows a heap-based buffer overflow (4-byte out-of-bounds write).

Oracle reports:

This Critical Patch Update contains 31 new security fixes for Oracle MySQL. 6 of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials.

The WebKitGTK project reports multiple vulnerabilities.

Gitlab reports:

Source branch of a MR could be removed by an unauthorised user

Private group members could be listed

Disclosure of System Notes via Elasticsearch integration

Disclosure of Private Comments via Elasticsearch integration

Confirm existence of private repositories

Private group membership could be disclosed

Disclosure of Project Labels

Disclosure of Private Project Path and Labels

Uncontrolled Resource Consumption due to Nested GraphQL Queries

Improper access control on comments

Sentry Token Access Control

Authorisation check for Project Transfer option

XSS in Wiki Pages Using RDoc

Untrusted Input could be used for Internal Redirect

Access control for protected environments

Private Sub Group path Disclosure

Disclosure of Group Packages List

Private Repository Name Disclosure

The Gitea Team reports:

When a comment in an issue or PR mentions a user using @username, the mentioned user receives a mail notification even if they don't have permission to see the originating repository.

Matrix developers report:

Make sure that [...] events sent over /send_join, /send_leave, and /invite, are correctly signed and come from the expected servers.

The samba project reports:

Malicious servers can cause Samba client code to return filenames containing path separators to calling code.

When the password contains multi-byte (non-ASCII) characters, the check password script does not receive the full password string.

Users with the "get changes" extended access right can crash the AD DC LDAP server by requesting an attribute using the range= syntax.

Varnish Software reports:

A bug has been discovered in Varnish Cache where we fail to clear a pointer between the handling of one client requests and the next on the same connection. This can under specific circumstances lead to information being leaked from the connection workspace.

Todd C. Miller reports:

When sudo is configured to allow a user to run commands as an arbitrary user via the ALL keyword in a Runas specification, it is possible to run commands as root by specifying the user ID -1 or 4294967295.

This can be used by a user with sufficient sudo privileges to run commands as root even if the Runas specification explicitly disallows root access as long as the ALL keyword is listed first in the Runas specification.

Log entries for commands run this way will list the target user as 4294967295 instead of root. In addition, PAM session modules will not be run for the command.

Problem Description:

The decompressor used in bzip2 contains a bug which can lead to an out-of-bounds write when processing a specially crafted bzip2(1) file.

bzip2recover contains a heap use-after-free bug which can be triggered when processing a specially crafted bzip2(1) file.

Impact:

An attacker who can cause maliciously crafted input to be processed may trigger either of these bugs. The bzip2recover bug may cause a crash, permitting a denial-of-service. The bzip2 decompressor bug could potentially be exploited to execute arbitrary code.

Note that some utilities, including the tar(1) archiver and the bspatch(1) binary patching utility (used in portsnap(8) and freebsd-update(8)) decompress bzip2(1)-compressed data internally; system administrators should assume that their systems will at some point decompress bzip2(1)-compressed data even if they never explicitly invoke the bunzip2(1) utility.

Problem Description:

The ICMPv6 input path incorrectly handles cases where an MLDv2 listener query packet is internally fragmented across multiple mbufs.

Impact:

A remote attacker may be able to cause an out-of-bounds read or write that may cause the kernel to attempt to access an unmapped page and subsequently panic.

Problem Description:

A function extracting the length from type-length-value encoding is not properly validating the submitted length.

Impact:

A remote user could cause, for example, an out-of-bounds read, decoding of unrelated data, or trigger a crash of the software such as bsnmpd resulting in a denial of service.

Problem Description:

The e1000 network adapters permit a variety of modifications to an Ethernet packet when it is being transmitted. These include the insertion of IP and TCP checksums, insertion of an Ethernet VLAN header, and TCP segmentation offload ("TSO"). The e1000 device model uses an on-stack buffer to generate the modified packet header when simulating these modifications on transmitted packets.

When TCP segmentation offload is requested for a transmitted packet, the e1000 device model used a guest-provided value to determine the size of the on-stack buffer without validation. The subsequent header generation could overflow an incorrectly sized buffer or indirect a pointer composed of stack garbage.

Impact:

A misbehaving bhyve guest could overwrite memory in the bhyve process on the host.

Problem Description:

Due do a missing check in the code of m_pulldown(9) data returned may not be contiguous as requested by the caller.

Impact:

Extra checks in the IPv6 code catch the error condition and trigger a kernel panic leading to a remote DoS (denial-of-service) attack with certain Ethernet interfaces. At this point it is unknown if any other than the IPv6 code paths can trigger a similar condition.

Problem Description:

The kernel driver for /dev/midistat implements a handler for read(2). This handler is not thread-safe, and a multi-threaded program can exploit races in the handler to cause it to copy out kernel memory outside the boundaries of midistat's data buffer.

Impact:

The races allow a program to read kernel memory within a 4GB window centered at midistat's data buffer. The buffer is allocated each time the device is opened, so an attacker is not limited to a static 4GB region of memory.

On 32-bit platforms, an attempt to trigger the race may cause a page fault in kernel mode, leading to a panic.

Problem Description:

System calls operating on file descriptors obtain a reference to relevant struct file which due to a programming error was not always put back, which in turn could be used to overflow the counter of affected struct file.

Impact:

A local user can use this flaw to obtain access to files, directories, sockets, etc., opened by processes owned by other users. If obtained struct file represents a directory from outside of user's jail, it can be used to access files outside of the jail. If the user in question is a jailed root they can obtain root privileges on the host system.

GitHub issue:

This issue has been created for public disclosure of an XSS vulnerability that was responsibly reported by https://hackerone.com/vxhex

In the Loofah gem, through v2.3.0, unsanitized JavaScript may occur in sanitized output when a crafted SVG element is republished.

Python changelog:

bpo-38243: Escape the server title of xmlrpc.server.DocXMLRPCServer when rendering the document page as HTML.

bpo-38174: Update vendorized expat library version to 2.2.8, which resolves CVE-2019-15903.

bpo-37764: Fixes email._header_value_parser.get_unstructured going into an infinite loop for a specific case in which the email header does not have trailing whitespace, and the case in which it contains an invalid encoded word.

bpo-37461: Fix an infinite loop when parsing specially crafted email headers.

bpo-34155: Fix parsing of invalid email addresses with more than one @ (e.g. a@b@c.com.) to not return the part before 2nd @ as valid email address.

Mitre reports:

An issue was discovered in Pillow before 6.2.0. When reading specially crafted invalid image files, the library can either allocate very large amounts of memory or take an extremely long period of time to process the image.

mod_perl2 2.0.11 fixes Arbitrary Perl code execution in the context of the user account via a user-owned .htaccess.

mod_perl 2.0 through 2.0.10 allows attackers to execute arbitrary Perl code by placing it in a user-owned .htaccess file, because (contrary to the documentation) there is no configuration option that permits Perl code for the administrator's control of HTTP request processing without also permitting unprivileged users to run Perl code in the context of the user account that runs Apache HTTP Server processes.

Xpdf 4.02 fixes two vulnerabilities. Both fixes have been backported to 3.04.

An invalid memory access vulnerability in TextPage::findGaps() in Xpdf 4.01 through a crafted PDF document can cause a segfault.

An out of bounds write exists in TextPage::findGaps() of Xpdf 4.01.01

Unbound Security Advisories:

Due to an error in parsing NOTIFY queries, it is possible for Unbound to continue processing malformed queries and may ultimately result in a pointer dereference in uninitialized memory. This results in a crash of the Unbound daemon.

Gitlab reports:

Disclosure of Private Code, Merge Requests and Commits via Elasticsearch integration

SO-AND-SO reports:

XSS in Markdown Preview Using Mermaid

Bypass Email Verification using Salesforce Authentication

Account Takeover using SAML

Uncontrolled Resource Consumption in Markdown using Mermaid

Disclosure of Private Project Path and Labels

Disclosure of Assignees via Milestones

Disclosure of Project Path via Unsubscribe Link

Disclosure of Project Milestones via Groups

Disclosure of Private System Notes via GraphQL

GIT Command Injection via API

Bypass User Blocking via CI/CD token

IDOR Adding Groups to Protected Environments

Disclosure of Group Membership via Merge Request Approval Rules

Disclosure of Head Pipeline via Blocking Merge Request Feature

Grafana update

The cacti developers reports:

In Cacti through 1.2.6, authenticated users may bypass authorization checks (for viewing a graph) via a direct graph_json.php request with a modified local_graph_id parameter.

Ruby news:

This release includes security fixes. Please check the topics below for details.

CVE-2019-15845: A NUL injection vulnerability of File.fnmatch and File.fnmatch?

A NUL injection vulnerability of Ruby built-in methods (File.fnmatch and File.fnmatch?) was found. An attacker who has the control of the path pattern parameter could exploit this vulnerability to make path matching pass despite the intention of the program author.

CVE-2019-16201: Regular Expression Denial of Service vulnerability of WEBrick's Digest access authentication

Regular expression denial of service vulnerability of WEBrick's Digest authentication module was found. An attacker can exploit this vulnerability to cause an effective denial of service against a WEBrick service.

CVE-2019-16254: HTTP response splitting in WEBrick (Additional fix)

There is an HTTP response splitting vulnerability in WEBrick bundled with Ruby.

CVE-2019-16255: A code injection vulnerability of Shell#[] and Shell#test

A code injection vulnerability of Shell#[] and Shell#test in a standard library (lib/shell.rb) was found.

Sicheng Liu of Beijing DBSEC Technology Co., Ltd reports:

Incorrect scoping of kill operations in MongoDB Server's packaged SysV init scripts allow users with write access to the PID file to insert arbitrary PIDs to be killed when the root user stops the MongoDB process via SysV init.

Rich Mirch reports:

An unprivileged user or program on Microsoft Windows which can create OpenSSL configuration files in a fixed location may cause utility programs shipped with MongoDB server versions less than 4.0.11, 3.6.14, and 3.4.22 to run attacker defined code as the user running the utility.

Exim developers team report:

There is a heap overflow in string_vformat().Using a EHLO message, remote code execution seems to be possible.

Mitch Wasson of Cisco's Advanced Malware Protection Group reports:

After user deletion in MongoDB Server the improper invalidation of authorization sessions allows an authenticated user's session to persist and become conflated with new accounts, if those accounts reuse the names of deleted ones.

The Mantis developers report:

CVE-2019-15715: [Admin Required - Post Authentication] Command Execution / Injection Vulnerability

CVE-2019-8331: In Bootstrap before 3.4.1, XSS is possible in the tooltip or popover data-template attribute

Missing integrity hashes for CSS resources from CDNs

The Go project reports:

net/http (through net/textproto) used to accept and normalize invalid HTTP/1.1 headers with a space before the colon, in violation of RFC 7230. If a Go server is used behind a reverse proxy that accepts and forwards but doesn't normalize such invalid headers, the reverse proxy and the server can interpret the headers differently. This can lead to filter bypasses or request smuggling, the latter if requests from separate clients are multiplexed onto the same connection by the proxy. Such invalid headers are now rejected by Go servers, and passed without normalization to Go client applications.

Jenkins Security Advisory:

Description

(Medium) SECURITY-1498 / CVE-2019-10401

Stored XSS vulnerability in expandable textbox form control

(Medium) SECURITY-1525 / CVE-2019-10402

XSS vulnerability in combobox form control

(Medium) SECURITY-1537 (1) / CVE-2019-10403

Stored XSS vulnerability in SCM tag action tooltip

(Medium) SECURITY-1537 (2) / CVE-2019-10404

Stored XSS vulnerability in queue item tooltip

(Medium) SECURITY-1505 / CVE-2019-10405

Diagnostic web page exposed Cookie HTTP header

(Medium) SECURITY-1471 / CVE-2019-10406

XSS vulnerability in Jenkins URL setting

Internet Systems Consortium, Inc. reports:

A packet containing a malformed DUID can cause the kea-dhcp6 server to terminate (CVE-2019-6472) [Medium]

An invalid hostname option can cause the kea-dhcp4 server to terminate (CVE-2019-6473) [Medium]

An oversight when validating incoming client requests can lead to a situation where the Kea server

will exit when trying to restart (CVE-2019-6474) [Medium]

Janos Follath reports:

Mbed TLS does not have a constant-time/constant-trace arithmetic library and uses blinding to protect against side channel attacks.

In the ECDSA signature routine previous Mbed TLS versions used the same RNG object for generating the ephemeral key pair and for generating the blinding values. The deterministic ECDSA function reused this by passing the RNG object created from the private key and the message to be signed as prescribed by RFC 6979. This meant that the same RNG object was used whenever the same message was signed, rendering the blinding ineffective.

If the victim can be tricked to sign the same message repeatedly, the private key may be recoverable through side channels.

Jon Siwek of Corelight reports:

This is a security patch release to address a potential Denial of Service vulnerability:

• The NTLM analyzer did not properly handle AV Pair sequences that were either empty or unterminated, resulting in invalid memory access or heap buffer over-read. The NTLM analyzer is enabled by default and used in the analysis of SMB, DCE/RPC, and GSSAPI protocols.

expat project reports:

Fix heap overflow triggered by XML_GetCurrentLineNumber (or XML_GetCurrentColumnNumber), and deny internal entities closing the doctype

expat project reports:

XML names with multiple colons could end up in the wrong namespace, and take a high amount of RAM and CPU resources while processing, opening the door to use for denial-of-service attacks

curl security problems:

CVE-2019-5481: FTP-KRB double-free

libcurl can be told to use kerberos over FTP to a server, as set with the CURLOPT_KRBLEVEL option.

During such kerberos FTP data transfer, the server sends data to curl in blocks with the 32 bit size of each block first and then that amount of data immediately following.

A malicious or just broken server can claim to send a very large block and if by doing that it makes curl's subsequent call to realloc() to fail, curl would then misbehave in the exit path and double-free the memory.

In practical terms, an up to 4 GB memory area may very well be fine to allocate on a modern 64 bit system but on 32 bit systems it will fail.

Kerberos FTP is a rarely used protocol with curl. Also, Kerberos authentication is usually only attempted and used with servers that the client has a previous association with.

CVE-2019-5482: TFTP small blocksize heap buffer overflow

libcurl contains a heap buffer overflow in the function (tftp_receive_packet()) that receives data from a TFTP server. It can call recvfrom() with the default size for the buffer rather than with the size that was used to allocate it. Thus, the content that might overwrite the heap memory is controlled by the server.

This flaw is only triggered if the TFTP server sends an OACK without the BLKSIZE option, when a BLKSIZE smaller than 512 bytes was requested by the TFTP client. OACK is a TFTP extension and is not used by all TFTP servers.

Users choosing a smaller block size than default should be rare as the primary use case for changing the size is to make it larger.

It is rare for users to use TFTP across the Internet. It is most commonly used within local networks. TFTP as a protocol is always inherently insecure.

This issue was introduced by the add of the TFTP BLKSIZE option handling. It was previously incompletely fixed by an almost identical issue called CVE-2019-5436.

Gitlab reports:

Project Template Functionality Could Be Used to Access Restricted Project Data

Security Enhancements in GitLab Pages

The OpenSSL project reports:

ECDSA remote timing attack (CVE-2019-1547) [Low]

Fork Protection (CVE-2019-1549) [Low]

(OpenSSL 1.1.1 only)

Adobe reports:

• This update resolves a same origin method execution vulnerability that could lead to arbitrary code execution (CVE-2019-8069).
• This update resolves a use-after-free vulnerability that could lead to arbitrary code execution (CVE-2019-8070).

Japheth Cleaver reports:

Several buffer overflows were reported by University of Cambridge Computer Security Incident Response Team.

A use-after-free in onig_new_deluxe() in regext.c in Oniguruma 6.9.2 allows attackers to potentially cause information disclosure, denial of service, or possibly code execution by providing a crafted regular expression. The attacker provides a pair of a regex pattern and a string, with a multi-byte encoding that gets handled by onig_new_deluxe().

A NULL Pointer Dereference in match_at() in regexec.c in Oniguruma 6.9.2 allows attackers to potentially cause denial of service by providing a crafted regular expression.

Oniguruma issues often affect Ruby, as well as common optional libraries for PHP and Rust.

Exim developers report:

If your Exim server accepts TLS connections, it is vulnerable. This does not depend on the TLS libray, so both, GnuTLS and OpenSSL are affected.

The vulnerability is exploitable by sending a SNI ending in a backslash-null sequence during the initial TLS handshake. The exploit exists as a POC. For more details see the document qualys.mbx

The Asterisk project reports:

When audio frames are given to the audio transcoding support in Asterisk the number of samples are examined and as part of this a message is output to indicate that no samples are present. A change was done to suppress this message for a particular scenario in which the message was not relevant. This change assumed that information about the origin of a frame will always exist when in reality it may not.

This issue presented itself when an RTP packet containing no audio (and thus no samples) was received. In a particular transcoding scenario this audio frame would get turned into a frame with no origin information. If this new frame was then given to the audio transcoding support a crash would occur as no samples and no origin information would be present. The transcoding scenario requires the genericplc option to be set to enabled (the default) and a transcoding path from the source format into signed linear and then from signed linear into another format.

Note that there may be other scenarios that have not been found which can cause an audio frame with no origin to be given to the audio transcoding support and thus cause a crash.

wordpress developers reports:

Props to Simon Scannell of RIPS Technologies for finding and disclosing two issues. The first, a cross-site scripting (XSS) vulnerability found in post previews by contributors. The second was a cross-site scripting vulnerability in stored comments.

Props to Tim Coen for disclosing an issue where validation and sanitization of a URL could lead to an open redirect.

Props to Anshul Jain for disclosing reflected cross-site scripting during media uploads.

Props to Zhouyuan Yang of Fortinets FortiGuard Labs who disclosed a vulnerability for cross-site scripting (XSS) in shortcode previews.

Props to Ian Dunn of the Core Security Team for finding and disclosing a case where reflected cross-site scripting could be found in the dashboard.

Props to Soroush Dalilifrom NCC Group for disclosing an issue with URL sanitization that can lead to cross-site scripting (XSS) attacks.

In addition to the above changes, we are also updating jQuery on older versions of WordPress. This change was added in 5.2.1 and is now being brought to older versions.

The Asterisk project reports:

When Asterisk sends a re-invite initiating T.38 faxing, and the endpoint responds with a declined media stream a crash will then occur in Asterisk.

Mozilla Foundation reports:

CVE-2019-11751: Malicious code execution through command line parameters

CVE-2019-11746: Use-after-free while manipulating video

CVE-2019-11744: XSS by breaking out of title and textarea elements using innerHTML

CVE-2019-11742: Same-origin policy violation with SVG filters and canvas to steal cross-origin images

CVE-2019-11736: File manipulation and privilege escalation in Mozilla Maintenance Service

CVE-2019-11753: Privilege escalation with Mozilla Maintenance Service in custom Firefox installation location

CVE-2019-11752: Use-after-free while extracting a key value in IndexedDB

CVE-2019-9812: Sandbox escape through Firefox Sync

CVE-2019-11741: Isolate addons.mozilla.org and accounts.firefox.com

CVE-2019-11743: Cross-origin access to unload event attributes

CVE-2019-11748: Persistence of WebRTC permissions in a third party context

CVE-2019-11749: Camera information available without prompting using getUserMedia

CVE-2019-5849: Out-of-bounds read in Skia

CVE-2019-11750: Type confusion in Spidermonkey

CVE-2019-11737: Content security policy directives ignore port and path if host is a wildcard

CVE-2019-11738: Content security policy bypass through hash-based sources in directives

CVE-2019-11747: 'Forget about this site' removes sites from pre-loaded HSTS list

CVE-2019-11734: Memory safety bugs fixed in Firefox 69

CVE-2019-11735: Memory safety bugs fixed in Firefox 69 and Firefox ESR 68.1

CVE-2019-11740: Memory safety bugs fixed in Firefox 69, Firefox ESR 68.1, and Firefox ESR 60.9

The samba project reports:

On a Samba SMB server for all versions of Samba from 4.9.0 clients are able to escape outside the share root directory if certain configuration parameters set in the smb.conf file.

The Varnish Team reports:

A failure in HTTP/1 parsing can allow a remote attacker to trigger an assertion in varnish, restarting the daemon and clearing the cache.

GnuPG reports:

Mitigate an ECDSA timing attack.

Ruby news:

There are multiple vulnerabilities about Cross-Site Scripting (XSS) in jQuery shipped with RDoc which bundled in Ruby. All Ruby users are recommended to update Ruby to the latest release which includes the fixed version of RDoc.

The following vulnerabilities have been reported.

CVE-2012-6708

CVE-2015-9251

Gitlab reports:

Kubernetes Integration Server-Side Request Forgery

Server-Side Request Forgery in Jira Integration

Improved Protection Against Credential Stuffing Attacks

Markdown Clientside Resource Exhaustion

Pipeline Status Disclosure

Group Runner Authorization Issue

CI Metrics Disclosure

User IP Disclosed by Embedded Image and Media

Label Description HTML Injection

IDOR in Epic Notes API

Push Rule Bypass

Project Visibility Restriction Bypass

Merge Request Discussion Restriction Bypass

Disclosure of Merge Request IDs

Weak Authentication In Certain Account Actions

Disclosure of Commit Title and Comments

Stored XSS via Markdown

EXIF Geolocation Data Exposure

Multiple SSRF Regressions on Gitaly

Default Branch Name Exposure

Potential Denial of Service via CI Pipelines

Privilege Escalation via Logrotate

The WebKitGTK project reports many vulnerabilities, including several arbitrary code execution vulnerabilities.

Mozilla Foundation reports:

CVE-2019-11733: Stored passwords in 'Saved Logins' can be copied without master password entry

When a master password is set, it is required to be entered again before stored passwords can be accessed in the 'Saved Logins' dialog. It was found that locally stored passwords can be copied to the clipboard thorough the 'copy password' context menu item without re-entering the master password if the master password had been previously entered in the same session, allowing for potential theft of stored passwords.

Jenkins Security Advisory:

Description

(Medium) SECURITY-1453 / CVE-2019-10383

Stored XSS vulnerability in update center

(High) SECURITY-1491 / CVE-2019-10384

CSRF protection tokens for anonymous users did not expire in some circumstances

Aki Tuomi reports:

Vulnerability Details: IMAP and ManageSieve protocol parsers do not properly handle NUL byte when scanning data in quoted strings, leading to out of bounds heap memory writes. Risk: This vulnerability allows for out-of-bounds writes to objects stored on the heap up to 8096 bytes in pre-login phase, and 65536 bytes post-login phase, allowing sufficiently skilled attacker to perform complicated attacks that can lead to leaking private information or remote code execution. Abuse of this bug is very difficult to observe, as it does not necessarily cause a crash. Attempts to abuse this bug are not directly evident from logs.

Jonathon Loomey of Netflix reports:

HTTP/2 implementations do not robustly handle abnormal traffic and resource exhaustion

Recently, a series of DoS attack vulnerabilities have been reported on a broad range of HTTP/2 stacks. Among the vulnerabilities, H2O is exposed to the following:

• CVE-2019-9512 "Ping Flood": The attacker sends continual pings to an HTTP/2 peer, causing the peer to build an internal queue of responses. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.
• CVE-2019-9514 "Reset Flood": The attacker opens a number of streams and sends an invalid request over each stream that should solicit a stream of RST_STREAM frames from the peer. Depending on how the peer queues the RST_STREAM frames, this can consume excess memory, CPU, or both, potentially leading to a denial of service.
• CVE-2019-9515 "Settings Flood": The attacker sends a stream of SETTINGS frames to the peer. Since the RFC requires that the peer reply with one acknowledgement per SETTINGS frame, an empty SETTINGS frame is almost equivalent in behavior to a ping. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.

Jonathon Loomey of Netflix reports:

HTTP/2 implementations do not robustly handle abnormal traffic and resource exhaustion

Recently, a series of DoS attack vulnerabilities have been reported on a broad range of HTTP/2 stacks. Among the vulnerabilities, H2O is exposed to the following:

• CVE-2019-9512 "Ping Flood": The attacker sends continual pings to an HTTP/2 peer, causing the peer to build an internal queue of responses. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.
• CVE-2019-9514 "Reset Flood": The attacker opens a number of streams and sends an invalid request over each stream that should solicit a stream of RST_STREAM frames from the peer. Depending on how the peer queues the RST_STREAM frames, this can consume excess memory, CPU, or both, potentially leading to a denial of service.
• CVE-2019-9515 "Settings Flood": The attacker sends a stream of SETTINGS frames to the peer. Since the RFC requires that the peer reply with one acknowledgement per SETTINGS frame, an empty SETTINGS frame is almost equivalent in behavior to a ping. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.

The Gitea Team reports:

This release contains two security fixes, so we highly recommend updating.

Node.js reports:

Node.js, as well as many other implementations of HTTP/2, have been found vulnerable to Denial of Service attacks. See https://github.com/Netflix/security-bulletins/blob/master/advisories/third-party/2019-002.md for more information.

Updates are now available for all active Node.js release lines, including Linux ARMv6 builds for Node.js 8.x (which had been delayed).

We recommend that all Node.js users upgrade to a version listed below as soon as possible.

Vulnerabilities Fixed

Impact: All versions of Node.js 8 (LTS "Carbon"), Node.js 10 (LTS "Dubnium"), and Node.js 12 (Current) are vulnerable to the following:

• CVE-2019-9511 "Data Dribble": The attacker requests a large amount of data from a specified resource over multiple streams. They manipulate window size and stream priority to force the server to queue the data in 1-byte chunks. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.
• CVE-2019-9512 "Ping Flood": The attacker sends continual pings to an HTTP/2 peer, causing the peer to build an internal queue of responses. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.
• CVE-2019-9513 "Resource Loop": The attacker creates multiple request streams and continually shuffles the priority of the streams in a way that causes substantial churn to the priority tree. This can consume excess CPU, potentially leading to a denial of service.
• CVE-2019-9514 "Reset Flood": The attacker opens a number of streams and sends an invalid request over each stream that should solicit a stream of RST_STREAM frames from the peer. Depending on how the peer queues the RST_STREAM frames, this can consume excess memory, CPU, or both, potentially leading to a denial of service.
• CVE-2019-9515 "Settings Flood": The attacker sends a stream of SETTINGS frames to the peer. Since the RFC requires that the peer reply with one acknowledgement per SETTINGS frame, an empty SETTINGS frame is almost equivalent in behavior to a ping. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.
• CVE-2019-9516 "0-Length Headers Leak": The attacker sends a stream of headers with a 0-length header name and 0-length header value, optionally Huffman encoded into 1-byte or greater headers. Some implementations allocate memory for these headers and keep the allocation alive until the session dies. This can consume excess memory, potentially leading to a denial of service.
• CVE-2019-9517 "Internal Data Buffering": The attacker opens the HTTP/2 window so the peer can send without constraint; however, they leave the TCP window closed so the peer cannot actually write (many of) the bytes on the wire. The attacker then sends a stream of requests for a large response object. Depending on how the servers queue the responses, this can consume excess memory, CPU, or both, potentially leading to a denial of service.
• CVE-2019-9518 "Empty Frames Flood": The attacker sends a stream of frames with an empty payload and without the end-of-stream flag. These frames can be DATA, HEADERS, CONTINUATION and/or PUSH_PROMISE. The peer spends time processing each frame disproportionate to attack bandwidth. This can consume excess CPU, potentially leading to a denial of service. (Discovered by Piotr Sikora of Google)

Micah Snyder reports:

• An out of bounds write was possible within ClamAV&s NSIS bzip2 library when attempting decompression in cases where the number of selectors exceeded the max limit set by the library (CVE-2019-12900). The issue has been resolved by respecting that limit.
• The zip bomb vulnerability mitigated in 0.101.3 has been assigned the CVE identifier CVE-2019-12625. Unfortunately, a workaround for the zip-bomb mitigation was immediately identified. To remediate the zip-bomb scan time issue, a scan time limit has been introduced in 0.101.4. This limit now resolves ClamAV's vulnerability to CVE-2019-12625.

The VLC project reports:

Security: * Fix a buffer overflow in the MKV demuxer (CVE-2019-14970) * Fix a read buffer overflow in the avcodec decoder (CVE-2019-13962) * Fix a read buffer overflow in the FAAD decoder * Fix a read buffer overflow in the OGG demuxer (CVE-2019-14437, CVE-2019-14438) * Fix a read buffer overflow in the ASF demuxer (CVE-2019-14776) * Fix a use after free in the MKV demuxer (CVE-2019-14777, CVE-2019-14778) * Fix a use after free in the ASF demuxer (CVE-2019-14533) * Fix a couple of integer underflows in the MP4 demuxer (CVE-2019-13602) * Fix a null dereference in the dvdnav demuxer * Fix a null dereference in the ASF demuxer (CVE-2019-14534) * Fix a null dereference in the AVI demuxer * Fix a division by zero in the CAF demuxer (CVE-2019-14498) * Fix a division by zero in the ASF demuxer (CVE-2019-14535)

SO-AND-SO reports:

nsd-checkzone in NLnet Labs NSD 4.2.0 has a Stack-based Buffer Overflow in the dname_concatenate() function in dname.c.

The Git community reports:

A carefully constructed commit object with a very large number of parents may lead to potential out-of-bounds writes or potential denial of service.

The ProgramData configuration file is always read for compatibility with Git for Windows and Portable Git installations. The ProgramData location is not necessarily writable only by administrators, so we now ensure that the configuration file is owned by the administrator or the current user.

nvd.nist.gov reports

X.Org xdm 1.1.10, 1.1.11, and possibly other versions, when performing authentication using certain implementations of the crypt API function that can return NULL, allows remote attackers to cause a denial of service (NULL pointer dereference and crash) by attempting to log into an account whose password field contains invalid characters, as demonstrated using the crypt function from glibc 2.17 and later with (1) the "!" character in the salt portion of a password field or (2) a password that has been encrypted using DES or MD5 in FIPS-140 mode.

SO-AND-SO reports:

SECURITY: CVE-2019-10081

mod_http2: HTTP/2 very early pushes, for example configured with "H2PushResource", could lead to an overwrite of memory in the pushing request's pool, leading to crashes. The memory copied is that of the configured push link header values, not data supplied by the client.

SECURITY: CVE-2019-9517

mod_http2: a malicious client could perform a DoS attack by flooding a connection with requests and basically never reading responses on the TCP connection. Depending on h2 worker dimensioning, it was possible to block those with relatively few connections.

SECURITY: CVE-2019-10098

rewrite, core: Set PCRE_DOTALL flag by default to avoid unpredictable matches and substitutions with encoded line break characters.

SECURITY: CVE-2019-10092

Remove HTML-escaped URLs from canned error responses to prevent misleading text/links being displayed via crafted links.

SECURITY: CVE-2019-10097

mod_remoteip: Fix stack buffer overflow and NULL pointer deference when reading the PROXY protocol header.

CVE-2019-10082

mod_http2: Using fuzzed network input, the http/2 session handling could be made to read memory after being freed, during connection shutdown.

Joe Cooper reports:

I've rolled out Webmin version 1.930 and Usermin version 1.780 for all repositories. This release includes several security fixes, including one potentially serious one caused by malicious code inserted into Webmin and Usermin at some point on our build infrastructure. We're still investigating how and when, but the exploitable code has never existed in our github repositories, so we've rebuilt from git source on new infrastructure (and checked to be sure the result does not contain the malicious code).

I don't have a changelog for these releases yet, but I wanted to announce them immediately due to the severity of this issue. To exploit the malicious code, your Webmin installation must have Webmin -> Webmin Configuration -> Authentication -> Password expiry policy set to Prompt users with expired passwords to enter a new one. This option is not set by default, but if it is set, it allows remote code execution.

This release addresses CVE-2019-15107, which was disclosed earlier today. It also addresses a handful of XSS issues that we were notified about, and a bounty was awarded to the researcher (a different one) who found them.

nghttp2 GitHub releases:

This release fixes CVE-2019-9511 "Data Dribble" and CVE-2019-9513 "Resource Loop" vulnerability in nghttpx and nghttpd. Specially crafted HTTP/2 frames cause Denial of Service by consuming CPU time. Check out https://github.com/Netflix/security-bulletins/blob/master/advisories/third-party/2019-002.md for details. For nghttpx, additionally limiting inbound traffic by --read-rate and --read-burst options is quite effective against this kind of attack.

CVE-2019-9511 "Data Dribble": The attacker requests a large amount of data from a specified resource over multiple streams. They manipulate window size and stream priority to force the server to queue the data in 1-byte chunks. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.

CVE-2019-9513 "Ping Flood": The attacker sends continual pings to an HTTP/2 peer, causing the peer to build an internal queue of responses. Depending on how efficiently this data is queued, this can consume excess CPU, memory, or both, potentially leading to a denial of service.

Apple reports:

• CVE-2019-8696 and CVE-2019-8675: SNMP buffer overflows.
• IPP buffer overflow.
• Memory disclosure in the scheduler.
• DoS issues in the scheduler.

The traefik project reports:

Update of dependency to go go1.12.8 resolves potential HTTP/2 denial of service in traefik.

Jesse Smith (upstream author of the doas program) reported:

Previous versions of "doas" transferred most environment variables, such as USER, HOME, and PATH from the original user to the target user. Passing these variables could cause files in the wrong path or home directory to be read (or written to), which resulted in potential security problems.

Many thanks to Sander Bos for reporting this issue and explaining how it can be exploited.

NGINX Team reports:

Several security issues were identified in nginx HTTP/2 implementation which might cause excessive memory consumption and CPU usage (CVE-2019-9511, CVE-2019-9513, CVE-2019-9516). The issues affect nginx compiled with the ngx_http_v2_module (not compiled by default) if the http2 option of the listen directive is used in a configuration file.