### Duplicate Advisory
This advisory has been withdrawn because it is a duplicate of GHSA-w6xc-g9qj-vp32. This link is maintained to preserve external references.
### Original Description
The safe_traversal module in uutils coreutils, which provides protection against Time-of-Check to Time-of-Use (TOCTOU) symlink races using file-descriptor-relative syscalls, is incorrectly limited to Linux targets. On other Unix-like systems such as macOS and FreeBSD, the utility fails to utilize these protections, leaving directory traversal operations vulnerable to symlink race conditions.
A logic error in the ln utility of the uutils coreutils Rust implementation causes it to reject source paths with non-UTF-8 filename bytes when using target-directory forms. Unlike GNU ln, which treats filenames as raw bytes, uutils coreutils enforces UTF-8 encoding, leading to failure to stat such files and a non-zero exit code. This results in local denial of service for automated scripts or system tasks handling valid non-UTF-8 filenames on Unix filesystems. The issue affects multiple versions prior to 0.10.0-1ubuntu1 and has a medium severity rating.
A logic error in the ln utility of uutils coreutils allows the utility to dereference a symbolic link target even when the --no-dereference (or -n) flag is explicitly provided. The implementation previously only honored the "no-dereference" intent if the --force (overwrite) mode was also enabled. This flaw causes ln to follow a symbolic link that points to a directory and create new links inside that target directory instead of treating the symbolic link itself as the destination. In environments where a privileged user or system script uses ln -n to update a symlink, a local attacker could manipulate existing symbolic links to redirect file creation into sensitive directories, potentially leading to unauthorized file creation or system misconfiguration.
This release of the Red Hat OpenShift distributed tracing platform (Tempo) provides security improvements and bug fixes.
Breaking changes:
* None.
Deprecations:
* None.
Technology Preview features:
* None.
Enhancements:
* None.
Bug fixes:
* gRPC-Go authorization bypass vulnerability fix: Previously, gRPC-Go was vulnerable to an authorization bypass attack. This issue occurred because the HTTP/2 :path pseudo-header was not properly validated. Remote attackers could send raw HTTP/2 frames with a malformed :path that omitted the mandatory leading slash to bypass defined security policies. With this update, gRPC-Go properly validates the :path pseudo-header and rejects malformed requests. As a result, attackers can no longer bypass security policies to gain unauthorized access to services or disclose information. For more information, see https://access.redhat.com/security/cve/cve-2026-33186.
* XPath component fix: Previously, the github.com/antchfx/xpath component did not properly handle certain Boolean XPath expressions. A remote attacker could submit a crafted expression that caused an infinite loop, resulting in 100% CPU utilization and a denial-of-service condition. With this update, the XPath component correctly processes Boolean expressions that evaluate to true. The system no longer enters an infinite loop when handling these expressions. For more information, see https://access.redhat.com/security/cve/cve-2026-4645.
* Go JOSE denial-of-service vulnerability fix: Previously, the Go JOSE library for handling JSON Web Encryption (JWE) objects was vulnerable to a denial-of-service (DoS) attack. This issue occurred because the application failed when decrypting a specially crafted JWE object that specified a key wrapping algorithm but contained an empty encrypted key field. With this update, Go JOSE properly validates the encrypted key field before decryption. As a result, the application no longer crashes when processing malformed JWE objects, and the service remains available to legitimate users. For more information, see https://access.redhat.com/security/cve/cve-2026-34986.
* Lodash _.template function fix: Previously, the lodash _.template function validated the variable option but did not validate options.imports key names. Both options passed values to the same code execution path. An attacker with the ability to control options.imports key names or pollute Object.prototype could exploit this gap to execute arbitrary code. With this update, lodash validates options.imports key names by using the same rules applied to the variable option. The _.template function rejects invalid key names and prevents code injection through this path. For more information, see https://access.redhat.com/security/cve/cve-2026-4800.
* Go crypto/x509 and crypto/tls packages fix: Previously, the Go standard library crypto/x509 and crypto/tls packages did not limit the number of intermediate certificates processed during certificate chain building. An attacker could provide an excessive number of intermediate certificates, causing the system to perform an uncontrolled amount of work and resulting in a denial-of-service condition. With this update, the packages limit the number of intermediate certificates accepted during certificate chain validation. The system rejects certificate chains that exceed this limit. For more information, see https://access.redhat.com/security/cve/cve-2026-32280.
* Go Root.Chmod function fix: Previously, the Root.Chmod function in the Go standard library internal/syscall/unix package had a race condition between checking and modifying a target file. An attacker could replace the target with a symbolic link after the check but before the operation completed, causing the permission change to apply to the linked file instead. This allowed an attacker to bypass directory restrictions and change permissions on unintended files. With this update, the Root.Chmod function prevents this race condition. The function no longer follows symbolic links that replace the target during execution. For more information, see https://access.redhat.com/security/cve/cve-2026-32282.
* Go crypto/x509 package fix: Previously, the Go crypto/x509 package applied excluded DNS constraints to wildcard Subject Alternative Names (SANs) in a case-sensitive manner. An attacker could bypass certificate validation by using a different case in the wildcard SAN than the excluded DNS constraint specified. This allowed the system to accept a malicious certificate that should have been rejected. With this update, the package applies DNS constraints case-insensitively when validating wildcard SANs. Certificate chain verification correctly rejects certificates that match excluded DNS constraints regardless of case. For more information, see https://access.redhat.com/security/cve/cve-2026-33810.
* Go crypto/tls component fix: Previously, the Go crypto/tls component did not re-validate certificates against updated certif
An improper access control vulnerability in the canonical-livepatch snap client prior to version 10.15.0 allows a local unprivileged user to obtain a sensitive, root-level authentication token by sending an unauthenticated request to the livepatchd.sock Unix domain socket. This vulnerability is exploitable on systems where an administrator has already enabled the Livepatch client with a valid Ubuntu Pro subscription. This token allows an attacker to access Livepatch services using the victim's credentials, as well as potentially cause issues to the Livepatch server.
pytest through 9.0.2 on UNIX relies on directories with the `/tmp/pytest-of-{user}` name pattern, which allows local users to cause a denial of service or possibly gain privileges.
Node.js is a software development platform for building fast and scalable network applications in the JavaScript programming language.
Security Fix(es):
* nodejs: Nodejs denial of service (CVE-2026-21637)
* minimatch: minimatch: Denial of Service via specially crafted glob patterns (CVE-2026-26996)
* undici: Undici: Denial of Service due to uncontrolled resource consumption (CVE-2026-2581)
* undici: Undici: HTTP header injection and request smuggling vulnerability (CVE-2026-1527)
* undici: undici: Denial of Service via unbounded memory consumption during WebSocket permessage-deflate decompression (CVE-2026-1526)
* undici: Undici: Denial of Service via invalid WebSocket permessage-deflate extension parameter (CVE-2026-2229)
* undici: Undici: HTTP Request Smuggling and Denial of Service due to duplicate Content-Length headers (CVE-2026-1525)
* undici: undici: Denial of Service via crafted WebSocket frame with large length (CVE-2026-1528)
* nghttp2: nghttp2: Denial of Service via malformed HTTP/2 frames after session termination (CVE-2026-27135)
* Node.js: Node.js: Denial of Service via malformed Internationalized Domain Name processing (CVE-2026-21712)
* Node.js: Node.js: Denial of Service due to crafted HTTP `__proto__` header (CVE-2026-21710)
* Node.js: Node.js: Information disclosure due to `fs.realpathSync.native()` bypassing filesystem read restrictions (CVE-2026-21715)
* nodejs: Node.js: Permission bypass allows unauthorized modification of file permissions and ownership via incomplete security fix. (CVE-2026-21716)
* Node.js: Node.js: Unauthorized inter-process communication due to missing Unix Domain Socket permission checks (CVE-2026-21711)
* Node.js: Node.js: Information disclosure via timing oracle in HMAC verification (CVE-2026-21713)
* Node.js: Node.js: Memory leak and Denial of Service via crafted HTTP/2 WINDOW_UPDATE frames (CVE-2026-21714)
* nodejs: v8: Node.js: Denial of Service via V8 string hashing mechanism due to predictable hash collisions (CVE-2026-21717)
For more details about the security issue(s), including the impact, a CVSS score, acknowledgments, and other related information, refer to the CVE page(s) listed in the References section.
Tmds.DBus provides .NET libraries for working with D-Bus from .NET. Tmds.DBus and Tmds.DBus.Protocol are vulnerable to malicious D-Bus peers. A peer on the same bus can spoof signals by impersonating the owner of a well-known name, exhaust system resources or cause file descriptor spillover by sending messages with an excessive number of Unix file descriptors, and crash the application by sending malformed message bodies that cause unhandled exceptions on the SynchronizationContext. This vulnerability is fixed in Tmds.DBus 0.92.0 and Tmds.DBus.Protocol 0.92.0 and 0.21.3.
LiquidJS is a Shopify / GitHub Pages compatible template engine in pure JavaScript. Prior to 10.25.3, for {% include %}, {% render %}, and {% layout %}, LiquidJS checks whether the candidate path is inside the configured partials or layouts roots before reading it. That check is path-based, not realpath-based. Because of that, a file like partials/link.liquid passes the directory containment check as long as its pathname is under the allowed root. If link.liquid is actually a symlink to a file outside the allowed root, the filesystem follows the symlink when the file is opened and LiquidJS renders the external target. So the restriction is applied to the path string that was requested, not to the file that is actually read. This matters in environments where an attacker can place templates or otherwise influence files under a trusted template root, including uploaded themes, extracted archives, mounted content, or repository-controlled template trees. This vulnerability is fixed in 10.25.3.
On Linux, if the target of Root.Chmod is replaced with a symlink while the chmod operation is in progress, Chmod can operate on the target of the symlink, even when the target lies outside the root. The Linux fchmodat syscall silently ignores the AT_SYMLINK_NOFOLLOW flag, which Root.Chmod uses to avoid symlink traversal. Root.Chmod checks its target before acting and returns an error if the target is a symlink lying outside the root, so the impact is limited to cases where the target is replaced with a symlink between the check and operation.
OpenPrinting CUPS is an open source printing system for Linux and other Unix-like operating systems. In versions 2.4.16 and prior, a use-after-free vulnerability exists in the CUPS scheduler (cupsd) when temporary printers are automatically deleted.cupsdDeleteTemporaryPrinters() in scheduler/printers.c calls cupsdDeletePrinter() without first expiring subscriptions that reference the printer, leaving cupsd_subscription_t.dest as a dangling pointer to freed heap memory. The dangling pointer is subsequently dereferenced at multiple code sites, causing a crash (denial of service) of the cupsd daemon. With heap grooming, this can be leveraged for code execution.
OpenPrinting CUPS is an open source printing system for Linux and other Unix-like operating systems. In versions 2.4.16 and prior, an integer underflow vulnerability in _ppdCreateFromIPP() (cups/ppd-cache.c) allows any unprivileged local user to crash the cupsd root process by supplying a negative job-password-supported IPP attribute. The bounds check only caps the upper bound, so a negative value passes validation, is cast to size_t (wrapping to ~2^64), and is used as the length argument to memset() on a 33-byte stack buffer. This causes an immediate SIGSEGV in the cupsd root process. Combined with systemd's Restart=on-failure, an attacker can repeat the crash for sustained denial of service.
OpenPrinting CUPS is an open source printing system for Linux and other Unix-like operating systems. In versions 2.4.16 and prior, in a network-exposed cupsd with a shared target queue, an unauthorized client can send a Print-Job to that shared PostScript queue without authentication. The server accepts a page-border value supplied as textWithoutLanguage, preserves an embedded newline through option escaping and reparse, and then reparses the resulting second-line PPD: text as a trusted scheduler control record. A follow-up raw print job can therefore make the server execute an attacker-chosen existing binary such as /usr/bin/vim as lp. At time of publication, there are no publicly available patches.
OpenPrinting CUPS is an open source printing system for Linux and other Unix-like operating systems. In versions 2.4.16 and prior, there is a heap-based buffer overflow in the CUPS scheduler when building filter option strings from job attribute. At time of publication, there are no publicly available patches.
OpenPrinting CUPS is an open source printing system for Linux and other Unix-like operating systems. In versions 2.4.16 and prior, the RSS notifier allows .. path traversal in notify-recipient-uri (e.g., rss:///../job.cache), letting a remote IPP client write RSS XML bytes outside CacheDir/rss (anywhere that is lp-writable). In particular, because CacheDir is group-writable by default (typically root:lp and mode 0770), the notifier (running as lp) can replace root-managed state files via temp-file + rename(). This PoC clobbers CacheDir/job.cache with RSS XML, and after restarting cupsd the scheduler fails to parse the job cache and previously queued jobs disappear. At time of publication, there are no publicly available patches.
A flaw in Node.js Permission Model network enforcement leaves Unix Domain Socket (UDS) server operations without the required permission checks, while all comparable network paths correctly enforce them.
As a result, code running under `--permission` without `--allow-net` can create and expose local IPC endpoints, allowing communication with other processes on the same host outside of the intended network restriction boundary.
This vulnerability affects Node.js **25.x** processes using the Permission Model where `--allow-net` is intentionally omitted to restrict network access. Note that `--allow-net` is currently an experimental feature.
flagd is a feature flag daemon with a Unix philosophy. Prior to 0.14.2, flagd exposes OFREP (/ofrep/v1/evaluate/...) and gRPC (evaluation.v1, evaluation.v2) endpoints for feature flag evaluation. These endpoints are designed to be publicly accessible by client applications. The evaluation context included in request payloads is read into memory without any size restriction. An attacker can send a single HTTP request with an arbitrarily large body, causing flagd to allocate a corresponding amount of memory. This leads to immediate memory exhaustion and process termination (e.g., OOMKill in Kubernetes environments). flagd does not natively enforce authentication on its evaluation endpoints. While operators may deploy flagd behind an authenticating reverse proxy or similar infrastructure, the endpoints themselves impose no access control by default. This vulnerability is fixed in 0.14.2.
Local privilege escalation due to insecure Unix socket permissions. The following products are affected: Acronis Cyber Protect 17 (macOS) before build 41186, Acronis Cyber Protect Cloud Agent (macOS) before build 41124, Acronis True Image (macOS) before build 42902.