Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, contain a vulnerability in the HTTP endpoint used for creating the first user account. This endpoint does not verify whether the setup wizard has been completed, leading to a race condition. An unauthenticated attacker can exploit this flaw to create an administrator account during the initial user account replacement process.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have a vulnerability in how symbolic links are checked during artifact extraction. The software only checks for symbolic links in the path before creating parent directories, not immediately before writing the extracted file. This flaw allows attackers with Item/Configure permission to write files to arbitrary locations on the Jenkins controller file system by exploiting symbolic links in artifacts.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have an HTTP endpoint that loads agents by name without validating the name. This flaw allows attackers with Overall/Read permission to load any config.xml file on the controller filesystem as an agent configuration and overwrite arbitrary config.xml files when saving the agent, potentially leading to arbitrary code execution.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have a vulnerability where symbolic links are followed when copying the directory of an item being renamed. This behavior allows attackers with Item/Create and Item/Configure permissions to read arbitrary files on the Jenkins controller file system. The issue is fixed in Jenkins 2.588.0 and LTS 2.580.2.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have a vulnerability where launch commands in older agent configurations are converted into launchers from the Command Agent Launcher Plugin without administrator approval. This flaw allows attackers with Agent/Configure permission to execute arbitrary code within the Jenkins controller JVM. Fixed versions 2.588.0 and 2.580.2 address this issue.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have a vulnerability where item types in collections are not restricted after deserialization. This flaw allows attackers with access to POST config.xml API endpoints to submit crafted config.xml files, potentially leading to stored cross-site scripting (XSS) vulnerabilities.
Jenkins versions 2.587 and earlier, including LTS 2.580.1 and earlier, have a vulnerability where the deserialization filter does not apply to classes referenced by the 'resolves-to' attribute during XStream deserialization. This flaw was fixed in Jenkins 2.588.0 and LTS 2.580.2.
A vulnerability in Keycloak's SAML HTTP-Redirect binding allows an attacker to bypass signature verification by exploiting the server's processing of percent-encoded parameter names. This flaw enables an attacker to submit a crafted, unsigned SAML message that the server accepts as valid, potentially leading to full account takeover of federated users. The issue affects Keycloak versions from 26.6.0 up to but not including 26.6.9. A security update fixing this vulnerability is available in version 26.6.9.
An authenticated Kafka principal without authorization can exploit a flaw in IncrementalAlterConfigs or AlterConfigs requests to read arbitrary broker files or environment variables. This occurs because configuration validation happens before authorization checks, allowing resolution of configuration variables that expose sensitive data. The issue affects Kafka versions before 4.2.2 and between 4.3.0 and 4.3.2. The vulnerability is mitigated by upgrading to versions 4.2.2, 4.3.2, or 4.4.0, which enforce authorization before configuration validation and restrict config-provider usage.
Apache Kafka contains a vulnerability in several deserialization paths where unbounded memory allocation can be triggered by crafted input declaring large sizes. This can cause an OutOfMemoryError that crashes Kafka Streams or consumer threads. The issue affects Kafka versions prior to 4.2.2, 4.3.2, and 4.4.0. The broker and cluster remain unaffected. The vulnerability is mitigated by upgrading to fixed versions and restricting write access to internal topics and consumer groups.
Apache Kafka versions prior to 4.2.2 and between 4.3.0 and 4.3.2 have a vulnerability in the parsing of flexible protocol request headers. An attacker can send a request with a large declared count of tagged fields, causing excessive memory allocation and potentially exhausting broker heap memory or causing process-wide garbage collection stalls. This can lead to denial of service on the broker. The issue occurs before authentication and authorization, so no ACL or quota is required to exploit it. Fixed in versions 4.2.2, 4.3.2, and 4.4.0.
Apache Kafka contains a vulnerability in its code-generated protocol readers where an attacker can cause the broker to allocate excessive heap memory by declaring an array length much larger than the actual request buffer size. This leads to an OutOfMemoryError and broker crash. The issue affects all Kafka API types with array fields and occurs before authorization checks, allowing unauthenticated attackers on unauthenticated listeners to exploit it. The vulnerability is fixed in Apache Kafka versions 4.2.2, 4.3.2, and 4.4.0.
Apache Kafka's PushTelemetry API (KIP-714) decompresses client metrics payloads without limiting the decompressed size, allowing an authenticated client to send a small compressed payload that expands to a very large size in memory. This can exhaust broker heap memory, causing OutOfMemoryError and potentially crashing or degrading broker availability. The vulnerability affects Kafka versions from 3.7.0 up to but not including 4.2.2, and from 4.3.0 up to but not including 4.3.1. Fixed versions 4.2.2, 4.3.1, and 4.4.0 enforce a decompression size limit equal to telemetry.max.bytes (default 1 MB), preventing excessive memory use.
Apache Kafka brokers using SASL/OAUTHBEARER authentication with the default JWT validator and without a configured JWKS endpoint URL or custom verification key resolver silently skip JWT signature verification. This allows an unauthenticated remote attacker to forge tokens with any principal, including superusers, gaining full cluster access. The issue affects Kafka versions 4.1.0 through before 4.2.2 and 4.3.0 through before 4.3.1. Fixed versions 4.2.2 and 4.3.1 enforce signature verification or fail to start if misconfigured.
Apache Kafka brokers using SASL/OAUTHBEARER with the bundled JWKS-backed validator do not enforce JWT issuer and audience validation by default prior to version 4.4.0. This means that unless explicitly configured, tokens signed by any issuer or lacking an audience claim may be accepted if their signature is valid. From version 4.4.0, brokers require explicit configuration of issuer and audience validation or can opt out with specific flags.
A vulnerability in Grafana allows a user with query access to an Azure Monitor data source to trigger server-side request forgery (SSRF) that discloses the Azure access token used by the data source. This issue affects multiple Grafana versions prior to fixed releases and can lead to unauthorized disclosure of sensitive credentials.
A denial of service vulnerability in Grafana allows a user with Viewer role to crash the server by submitting a crafted server-side math expression that causes uncontrolled recursion. Users with permission to provision alert rules can cause the outage to persist across server restarts. The issue affects multiple Grafana versions prior to fixed releases in 12.4.14, 13.0.12, 13.1.9, and 13.2.5.
Grafana versions prior to 12.4.11, and certain 13.x versions before 13.0.9, 13.1.6, and 13.2.2, improperly log OAuth and OIDC tokens unredacted when tokens are invalid or missing. This allows users with read access to server logs or log stores to recover sensitive OAuth and OIDC access, refresh, and ID tokens. The exposure is limited to the scopes granted to the OAuth client.
CVE-2026-33379 is a permission bypass vulnerability in Grafana OSS and Grafana Enterprise version 12.4. It allows any user role, including Viewer or service accounts with no permissions, to create, externally publish, and delete dashboard snapshots due to missing authorization checks. This issue affects versions from 12.4.0 up to but not including 12.4.4. The vulnerability has been fixed in Grafana 12.4.4. Earlier versions and Grafana 13 are not affected.
Grafana's RBAC service had a flaw where users with dashboard permission-management rights on any folder could be incorrectly treated as administrators of all folders when viewing deleted dashboard listings. This allowed exposure of deleted dashboard items from folders without proper administrative access. The issue is fixed in version 13.1.2 by requiring appropriate folder-level permissions.