General Security News (unfiltered)
Scoped Vulnerability Tracking
The most urgent threat is a critical unauthenticated SQL injection vulnerability in Metacat (CVE-2026-48528) that allows for full data exfiltration and compromise of the underlying PostgreSQL database. Additionally, multiple high-severity vulnerabilities in MySQL and PostgreSQL require immediate attention, including a potential privilege escalation in MySQL via the Data Science Pipelines Operator and a client-side denial of service in the pgjdbc driver.
Summary: An unauthenticated SQL injection vulnerability exists in Metacat's REST API endpoints (`/cn/v1/object` and `/cn/v2/object`) due to unsanitized `nodeId` parameters. Attackers can use error-based techniques to extract arbitrary data or execute arbitrary SQL statements to modify or delete database contents.
Affected Features / Interoperability: Impacts the Metacat data repository software; specifically, the `/cn` REST API endpoints.
Source: https://github.com/NCEAS/metacat/security/advisories/GHSA-6g6j-wh5h-77h5
Actions:
Summary: A vulnerability in the `libpq` functions `lo_export()`, `lo_read()`, `lo_lseek64()`, and `lo_tell64()` allows a server superuser to overwrite a client stack buffer with an arbitrarily large response. This can lead to memory corruption in client tools like `psql` or `pg_dump`.
Affected Features / Interoperability: Impacts `libpq` functions and client-side tools like `psql` and `pg_dump`.
Source: https://www.postgresql.org/support/security/CVE-2026-6477/
Actions:
Summary: The `DataRow.Decode` function fails to validate field lengths, allowing a malicious or compromised server to send a message with a negative field length. This triggers a slice bounds out of range panic in the client.
Affected Features / Interoperability: Impacts the client-side decoding of DataRow messages.
Source: https://github.com/advisories/GHSA-jqcq-xjh3-6g23
Actions:
Summary: The `pgjdbc` driver is vulnerable to a client-side denial of service during SCRAM-SHA-256 authentication. A malicious server can provide a very large iteration count for PBKDF2, causing the client to consume unbounded CPU time.
Affected Features / Interoperability: Impacts the `pgjdbc` driver and can exhaust client CPU and connection pools.
Source: https://github.com/pgjdbc/pgjdbc/releases/tag/REL42.7.11
Actions:
Summary: TypeORM's `migration:generate` command fails to escape `${...}` interpolation in template literals. An attacker with database schema write access can inject payloads into column `COMMENT` or `DEFAULT` metadata that execute when migrations are run.
Affected Features / Interoperability: Impacts TypeORM migrations and schema metadata handling.
Source: https://github.com/typeorm/typeorm/releases/tag/0.3.31
Actions:
Summary: A vulnerability in the Data Science Pipelines Operator (DSPO) allows a namespace editor to inject dangerous parameters into the MySQL Data Source Name (DSN) string via the `spec.database.customExtraParams` field. This enables `LOCAL INFILE` functionality to exfiltrate sensitive files (like service account tokens) from the operator pod, leading to cluster-admin privileges.
Affected Features / Interoperability: Impacts the Data Science Pipelines Operator (DSPO) and Kubernetes cluster security.
Source: https://access.redhat.com/errata/RHSA-2026:53261
Actions:
Summary: In Budibase (prior to 3.39.18), the MySQL integration enabled `multipleStatements` and failed to escape table names in `DESCRIBE` statements. An attacker creating a table with a malicious name can trigger arbitrary statement execution when an administrator runs schema discovery.
Affected Features / Interoperability: Impacts Budibase MySQL integration and schema discovery functionality.
Source: https://github.com/Budibase/budibase/releases/tag/3.39.18
Actions:
Summary: TypeORM's `migration:generate` command fails to escape `${...}` interpolation in template literals. An attacker with database schema write access can inject payloads into column `COMMENT` or `DEFAULT` metadata that execute when migrations are run.
Affected Features / Interoperability: Impacts TypeORM migrations and schema metadata handling.
Source: https://github.com/typeorm/typeorm/releases/tag/0.3.31
Actions:
Summary: TypeORM's `migration:generate` command fails to escape `${...}` interpolation in template literals. An attacker with database schema write access can inject payloads into column `COMMENT` or `DEFAULT` metadata that execute when migrations are run.
Affected Features / Interoperability: Impacts TypeORM migrations and schema metadata handling.
Source: https://github.com/typeorm/typeorm/releases/tag/0.3.31
Actions:
Summary: A remote code execution vulnerability exists in Microsoft SQL Server Reporting Services (SSRS) due to incorrect handling of page requests.
Affected Features / Interoperability: Impacts Microsoft SQL Server Reporting Services.
Source: https://portal.msrc.microsoft.com/en-US/security-guidance/advisory/CVE-2020-0618
Actions:
Summary: Budibase (before 3.40.0) fails to redact datasource credentials in `STRING` typed fields. Authenticated users with table read permissions can retrieve MongoDB connection strings and Firebase private keys in plaintext.
Affected Features / Interoperability: Impacts Budibase datasource credential storage and read API.
Source: https://github.com/Budibase/budibase/security/advisories/GHSA-6mpp-gfg5-x2vv
Actions:
Summary: Budibase (before 3.40.0) contains a NoSQL injection vulnerability in the MongoDB datasource integration. Attackers can inject MongoDB operators through query parameters to bypass access controls, read arbitrary documents, or execute JavaScript via `$where`.
Affected Features / Interoperability: Impacts Budibase MongoDB datasource integration and access controls.
Source: https://github.com/Budibase/budibase/security/advisories/GHSA-pmpg-2mxq-6xwr
Actions:
Summary: Mongoose (prior to 6.13.10, 7.8.10, 8.24.1, and 9.7.2) is vulnerable to prototype pollution via `updateOne` when using user-controlled updates. This can set properties on `Object.prototype`, impacting application integrity and availability.
Affected Features / Interoperability: Impacts Mongoose object modeling and application integrity.
Source: https://github.com/Automattic/mongoose/security/advisories/GHSA-664h-wqgq-64gw
Actions:
PostgreSQL
Oracle Database
MySQL
Microsoft SQL Server
Redis
The most urgent issue is CVE-2026-14450 in Red Hat's MaaS API, a critical flaw allowing any pod in the cluster to bypass Kuadrant AuthPolicy gateway by forging HTTP headers, enabling unauthorized access, privilege escalation, and cross-tenant ServiceAccount token minting. This vulnerability is actively exploited in the wild according to CISA KEV, making immediate remediation a top priority.
Summary: A flaw in the MaaS API allows any pod within the cluster to bypass the Kuadrant AuthPolicy gateway by forging HTTP headers (`X-MaaS-Username` and `X-MaaS-Group`), which are trusted verbatim. This lack of first-party authentication enables an attacker to gain unauthorized access and escalate privileges, including minting Kubernetes ServiceAccount tokens in other tenants' namespaces, revoking API keys, and exfiltrating sensitive model access configuration.
Affected Features / Interoperability: Impacts the MaaS API and Kuadrant AuthPolicy gateway. Downstream effects include unauthorized access to tenant namespaces, compromised ServiceAccount tokens, and potential exposure of model access configurations.
Source:
Actions:
Summary: OpenChoreo versions prior to 1.0.2 and 1.1.2 serve caller-facing management APIs on an externally reachable agent listener without authentication, allowing network-reachable attackers to invoke `/api/proxy/` and `/api/exec/` operations, proxy the data-plane Kubernetes API, and execute commands in workload pods in multi-cluster deployments.
Affected Features / Interoperability: Affects OpenChoreo's cluster-gateway server and management APIs. Impacts multi-cluster deployments, workload pod security, and data-plane Kubernetes API proxying.
Source:
Actions:
Summary: Vault Secrets Operator versions 1.3.0 up to 1.4.1 are vulnerable to arbitrary file read and credential exfiltration via AppRole authentication configuration, allowing a tenant with limited Kubernetes RBAC permissions to read files from the operator pod's filesystem and transmit their contents to a tenant-controlled endpoint, potentially leading to privilege escalation.
Affected Features / Interoperability: Impacts Vault Secrets Operator's AppRole authentication and Kubernetes RBAC integration. Affects credential management and tenant isolation within the cluster.
Source:
Actions:
Summary: The Red Hat OpenShift AI (RHOAI) overlay for the training operator incorrectly aggregates `trainjobs` management permissions into the native Kubernetes `edit ClusterRole`, allowing any user with `edit ClusterRole` permissions in a namespace to create, modify, and delete `TrainJobs`. Combined with unrestricted PodTemplateSpec passthrough, this can lead to privilege escalation and arbitrary code execution.
Affected Features / Interoperability: Affects RHOAI training operator and Kubernetes RBAC. Impacts TrainJob management, service account impersonation, and pod configuration controls.
Source:
Actions:
Summary: The RHOAI training-operator allows users with standard edit or admin roles to escalate privileges by creating training jobs that impersonate service accounts, access the host filesystem, and execute arbitrary code remotely due to aggregation of training job permissions onto native Kubernetes edit and admin ClusterRoles with unrestricted PodTemplateSpec passthrough.
Affected Features / Interoperability: Impacts RHOAI training-operator and Kubernetes RBAC. Affects service account impersonation, host filesystem access, and pod template configurations.
Source:
Actions:
Summary: OpenChoreo versions from 1.2.0-rc.1 until 1.2.0 authorize `component:exec` and `wirelogs:view` using the caller-supplied project query parameter instead of `comp.Spec.Owner.ProjectName`, allowing a user with a project-scoped grant to execute commands in and read wirelogs from components owned by other projects in the same namespace.
Affected Features / Interoperability: Affects OpenChoreo's component execution and wirelog viewing APIs. Impacts multi-project isolation and command execution controls within shared namespaces.
Source:
Actions:
Summary: The Data Science Pipelines Operator (DSPO) ClusterRole includes excessive privileges such as executing commands within pods and managing cluster-wide roles. If the DSPO pod is compromised, an attacker could leverage these privileges to gain full administrative control over the Kubernetes cluster.
Affected Features / Interoperability: Impacts DSPO ClusterRole permissions and Kubernetes cluster administration. Affects pod exec capabilities and cluster-wide role management.
Source:
Actions:
Summary: Fulcio versions through 1.8.5 improperly follow cross-host redirects and attach Kubernetes ServiceAccount tokens during OIDC discovery, allowing a malicious or compromised issuer to perform blind SSRF, substitute and cache malicious JWKS keys, or disclose ServiceAccount tokens to external hosts.
Affected Features / Interoperability: Affects Fulcio's OIDC discovery and certificate issuance process. Impacts Kubernetes ServiceAccount token security and JWKS key management.
Source:
Actions:
Summary: OpenChoreo versions prior to 1.0.4, 1.1.4, and 1.2.1 hardcode `backend.auth.dangerouslyDisableDefaultAuthPolicy` and `auth.providers.guest.dangerouslyAllowOutsideDevelopment` to true, exposing `/api/*` without authentication and allowing unauthenticated catalog reads, scaffolder log reads, and catalog location creation or deletion.
Affected Features / Interoperability: Affects OpenChoreo Backstage backend authentication and API access controls. Impacts catalog management and scaffolder operations.
Source:
Actions:
Summary: The Feast operator allows a malicious tenant to inject arbitrary code into their feature repository, which is executed by an automated process with elevated privileges, allowing the tenant to steal sensitive credentials and potentially gain administrative control over the Kubernetes cluster.
Affected Features / Interoperability: Impacts Feast operator's feature repository processing and tenant isolation. Affects credential security and cluster administration.
Source:
Actions:
Summary: Moby versions prior to 29.5.1 and moby/moby v2 prior to v2.0.0-beta.14 resolve decompression binaries from the container's filesystem rather than the host's during archive uploads, allowing a malicious container image with trojanized binaries to achieve arbitrary code execution with full daemon privileges when a user uploads a compressed archive.
Affected Features / Interoperability: Affects Docker daemon archive handling and container image trust. Impacts `docker cp` operations and container runtime security.
Source:
Actions:
Summary: Moby versions prior to 29.3.1 have a flaw in plugin privilege validation during `docker plugin install`, where the daemon may incorrectly accept a privilege set that differs from the one approved by the user, potentially allowing plugins to gain unintended privileges.
Affected Features / Interoperability: Affects Docker plugin installation and privilege management. Impacts plugin security and system integrity.
Source:
Actions:
Summary: Flowise versions <= 3.1.2 contain a sandbox escape in the vm2/@flowiseai/nodevm JavaScript sandbox, allowing an authenticated user to escape via puppeteer.launch() parameters and execute arbitrary OS commands as the Flowise process user (root in Docker image).
Affected Features / Interoperability: Affects Flowise's custom function endpoint and JavaScript sandbox. Impacts container security and host file disclosure.
Source:
Actions: