Just published.
The newest CVEs, straight from CVE.org and NVD.
-
CVE-2026-118626
MLflow Tracking Server Issues Missing Authorization Privilege Escalation Vulnerability. This vulnerability allows remote attackers to escalate privileges on affected installations of MLflow. Authentication is required to exploit this vulnerability. The specific flaw exists within the configuration of the BEFORE_REQUEST_HANDLERS. The issue results from the lack of authorization prior to allowing access to functionality. An attacker can leverage this vulnerability to escalate privileges to resources normally protected from the user. Was ZDI-CAN-32108.
-
CVE-2026-114551
MailEnable Auth Cleartext Storage Information Disclosure Vulnerability. This vulnerability allows remote attackers to disclose sensitive information on affected installations of MailEnable. Authentication is not required to exploit this vulnerability. The specific flaw exists within the handling of mailbox credentials. The issue results from storing sensitive information in plaintext. An attacker can leverage this in conjunction with other vulnerabilities to disclose stored credentials, leading to further compromise. Was ZDI-CAN-34478.
-
CVE-2026-121729
emailflare before commit cf31f33 contains a missing authorization vulnerability in the inbox service /api/keys routes that allows any authenticated member to list, create and revoke API keys. Attackers can mint a global live API key accepted by /v1/send to send mail from any domain, or revoke existing keys to disable integrations.
-
CVE-2026-121730
emailflare before commit cf31f33 contains a missing authorization vulnerability in POST /api/inbox/compose that allows authenticated members to send mail as any sender address. Attackers can supply an arbitrary from field, which is passed unchecked to the Cloudflare send call, to impersonate other users or the organization to external recipients.
-
CVE-2026-121535
A vulnerability was found in miroslavpejic85 MiroTalk P2P up to 2.0.87. This vulnerability affects the function checkPassword of the file app/src/server.js of the component Room Action Handler. The manipulation results in improper restriction of excessive authentication attempts. It is possible to launch the attack remotely. Attacks of this nature are highly complex. It is stated that the exploitability is difficult. The exploit has been made public and could be used. Upgrading to version 2.1.38 is able to resolve this issue. The patch is identified as 55bec1d05f931d2e43633d6d0b7274f79ec719dd. Upgrading the affected component is advised.
-
CVE-2026-121539
A vulnerability was found in miroslavpejic85 MiroTalk P2P up to 2.0.87/2.1.38. Affected by this issue is the function CheckPassword of the file app/src/server.js of the component Socket IO Handler. Performing a manipulation of the argument peer_id results in improper access controls. It is possible to initiate the attack remotely. Upgrading to version 2.1.39 can resolve this issue. The patch is named 4e40471abc48bfc32b4b174b8844716dfc5c1471. You should upgrade the affected component.
-
CVE-2026-118237
An out-of-bounds read vulnerability was identified in GitHub Enterprise Server in the Git service (babeld). When proving possession of a host key, babeld's SSH implementation sized the session identifier by the current digest length rather than its own. An attacker with an account and SSH access could re-key to an algorithm with a longer hash, causing adjacent process memory to be signed and returned to the client, where a limited amount of it may be recoverable. This vulnerability affected all releases prior to 3.23 and was fixed in versions 3.22.3, 3.21.8, 3.20.10, and 3.19.14. This vulnerability was reported via the GitHub Bug Bounty program.
-
CVE-2026-104608
A remote code execution vulnerability was identified in GitHub Enterprise Server in the Git service (babeld). babeld held references into internal collections while calling into the SSH library, where a re-entrant callback could modify a collection and invalidate those references. An attacker with an account and SSH access to the instance could leverage this use-after-free into a controlled write and run arbitrary commands as babeld, including disclosure of its environment secrets. Exploitation required per-build memory offsets; failed attempts crashed babeld. This affected all releases prior to 3.23 and was fixed in 3.22.3, 3.21.8, 3.20.10, and 3.19.14. This vulnerability was reported via the GitHub Bug Bounty program.
-
CVE-2026-108327
Tanium addressed a path traversal vulnerability in Tanium Data Service.
-
CVE-2026-22058
StorageGRID (formerly StorageGRID Webscale) versions 11.3 and higher allow an attacker with product-specific knowledge and the ability to communicate with the Grid Network on certain ports to make unauthorized requests.
-
CVE-2026-22060
All versions of StorageGRID (formerly StorageGRID Webscale) in default & common configurations allow an unauthenticated attacker to access a limited set of APIs that are intended for use within the grid.
-
CVE-2026-121541
In Docker Buildx versions v0.19.0 through v0.38.0, an untrusted Bake definition that builds multiple targets together may expose files accessible to the Buildx client without the expected fs.read approval. Under certain conditions, filesystem paths from some targets are not verified against the approved entitlements. Users who run untrusted Bake definitions may be affected. The issue is fixed in v0.38.1.
-
CVE-2026-121542
In Docker Buildx versions v0.19.0 through v0.38.0, an untrusted Bake definition may expose files accessible to the Buildx client without the expected fs.read approval. Under certain conditions, paths under the same parent directory as the current working directory whose names use the working directory's basename as a prefix can skip the filesystem entitlement check. Exploitation requires specific matching filenames next to the current project directory. Users who run untrusted Bake definitions may be affected. The issue is fixed in v0.38.1.
-
CVE-2026-121614
Docker Sandboxes versions 0.31.0 before 0.50.0 do not neutralize repository-controlled Git configuration when running certain host-side Git commands. An attacker who can modify a host repository's Git configuration, for example through a direct-mode sandbox with the repository mounted read-write, can configure core.fsmonitor or another Git setting that invokes an external program. Git can execute the program during a later clone-mode fetch or, starting in version 0.39.0, an sbx kit push initiated by the user. The program runs on the host with the invoking user's permissions, outside the sandbox's microVM isolation. The read-only host repository mount provided to a clone-mode sandbox does not itself allow that sandbox to modify the host repository's Git configuration. The issue is fixed in version 0.50.0.
-
CVE-2026-121580
Server-side request forgery in the Jira ticketing filter lookup in Devolutions Server 2026.2.22.0 and earlier allows an authenticated low-privileged user to obtain the stored Jira service account credentials via an attacker-controlled server address in the request.
-
CVE-2026-120715
Server-side request forgery in the Active Directory credential validation in Devolutions Server 2026.3.12.0 and earlier allows an authenticated low-privileged user to capture the NetNTLMv2 hash of the DVLS service account via an attacker-controlled domain name in the request.
-
CVE-2026-120716
Insertion of sensitive information into sent data in the identity provider user synchronization in Devolutions Server 2026.3.13.0 and earlier allows an authenticated user with the delegated permission to update users from the identity provider to obtain other users' MFA secrets via the synchronization API response.
-
CVE-2026-120733
Cleartext storage of sensitive information in the user account data in Devolutions Server 2026.3.12.0 and earlier allows an attacker with read access to the database or a database backup to obtain users' TOTP shared secrets via the unencrypted copy of the security key written when an administrator saves a user.
-
CVE-2026-120734
Cleartext transmission of sensitive information in the email configuration test in Devolutions Server 2026.3.13.0 and earlier allows an authenticated user with access to the test action to obtain the stored SMTP password via a request that names an attacker-controlled SMTP host.
-
CVE-2026-120959
A memory-handling flaw in the MongoDB Atlas SQL ODBC Driver can cause values returned for a query to be written outside the bounds of a buffer provided by the calling application. A party able to influence the values a query returns may cause data to be placed at an unintended location in the client application's memory, which may result in unintended modification of that memory or termination of the client process. Reaching this condition depends on a specific, application-chosen way of requesting several result rows at once.
-
CVE-2026-115732
Incorrect authorization in the certificate-name matching of trapperkeeper-authorization, as used by Puppet Server, in the affected products, allows a client that holds a CA-signed certificate whose name extends or shortens an authorized hostname to be authorized as that host. This enables impersonation of other nodes and, in Puppet Enterprise, access to certificate-authority administration and remote command execution on managed nodes. Exploitation requires a certificate with a colliding name, which the certificate authority issues by approval or autosign.
-
CVE-2026-119636
Improper release of expired privileged access checkouts in the PAM scheduled check-in task in Devolutions Server 2025.3.2.0 through 2026.2.22.0 allows an authenticated user with checkout rights on a PAM account to retain access to that account on the target system after the checkout expires via a backlog of 21 or more newer active checkouts that prevents the expired checkout from being checked in and its password from being rotated.
-
CVE-2026-115731
Deserialization of untrusted data in the report-processing code of Puppet, as used by Puppet Server, in the affected releases, allows a client that holds any CA-signed agent certificate to execute arbitrary code on the Puppet Server as its service user by submitting a report containing a crafted rich-data type via PUT request. Where the certificate authority autosigns requests without strong verification, an unauthenticated network attacker can obtain such a certificate.
-
CVE-2026-63476
isomorphic-git is a pure JavaScript implementation of Git for Node.js and browsers. Prior to 1.38.6, isomorphic-git reads a server-controlled targetSize while applying a packfile delta in src/utils/applyDelta.js and passes it without a bound to Buffer.alloc(targetSize). A malicious or man-in-the-middle-controlled remote can send one delta header that causes git.clone() or git.fetch() to allocate approximately 2 GB before validating the actual delta output. GitPackIndex.fromPack swallows the later size-mismatch error, so a host with sufficient memory may retain the resource waste without a visible failure, while containers, serverless functions, and small virtual machines may terminate. Node.js heap limits do not mitigate the off-heap Buffer allocation. This issue is fixed in version 1.38.6.
-
CVE-2026-63489
Yeti bridges the gap between CTI and DFIR practitioners by providing a Forensics Intelligence platform and pipeline. In version 2.5.0 and earlier, YETI directly interpolates user-controlled sorting fields and filter_aliases attribute paths into ArangoDB AQL identifier and expression positions in core/database_arango.py. A low-privileged authenticated user can send crafted values to POST /api/v2/observables/search, POST /api/v2/entities/search, POST /api/v2/indicators/search, POST /api/v2/dfiq/search, POST /api/v2/search/, or POST /api/v2/graph/search, causing attacker-controlled AQL subexpressions to be parsed and evaluated. This can provide an error oracle, evaluate expressions against internal collections such as users, alter intended search logic, enumerate schema information, and consume excessive database resources. The verified injection points are read and search contexts, so arbitrary database writes are not established. No fixed version is available as of this review.
-
CVE-2026-63475
isomorphic-git is a pure JavaScript implementation of Git for Node.js and browsers. Prior to 1.38.6, isomorphic-git persists server-supplied circular symref= capabilities through updateRemoteRefs, and GitRefManager.resolve() follows the resulting loose symbolic references without a default depth limit or cycle detection. After a clone from a malicious or man-in-the-middle-controlled remote, operations such as log, status, checkout, and branch that resolve refs/remotes/origin/a or refs/remotes/origin/b do not terminate. The circular references remain on disk, leaving the repository unusable for the application and causing a persistent denial of service. This issue is fixed in version 1.38.6.
-
CVE-2026-120567
The MongoDB Entity Framework Core provider does not neutralize special elements in values passed to SetProperty when an ExecuteUpdate call also contains a setter that reads the entity being updated. In that case the update is sent as an aggregation pipeline, and a caller-supplied string beginning with $ is evaluated as an expression rather than stored as data. A party who can influence such a value may modify or remove fields in the records being updated, or copy other fields of the same record into them.
-
CVE-2026-82976
A flaw was found in the image-customization-controller used in OpenShift's bare metal provisioning stack. The controller registers an HTTP file server on port 8084 with no authentication. The custom imageFileSystem implementation exposes a browsable HTML directory listing of all registered images at the root path. The only access control on per-host images is the random UUID filename; the directory listing discloses those UUIDs to any caller on the provisioning network. Each served image embeds an Ignition config containing the base64-encoded cluster pull secret, optional IRONIC_RAMDISK_SSH_KEY, proxy settings, and per-host NetworkManager keyfiles. An adjacent network attacker can enumerate all image UUIDs via directory listing, download any image, extract the embedded Ignition config, and recover the cluster pull secret and SSH key. ### Evidence **pkg/imagehandler/imagefilesystem.go — root directory is browsable and enumerates all UUID image names** ```go func (f *imageFileSystem) Readdir(n int) ([]fs.FileInfo, error) { f.mu.Lock() defer f.mu.Unlock() result := []fs.FileInfo{} for _, im := range f.images { result = append(result, im) } return result, nil } func (f *imageFileSystem) Open(name string) (http.File, error) { f.log.Info("Open", "path", name) if name == "/" { return f, nil } ... ``` **cmd/controller/main.go — file server bound on all interfaces without auth or TLS** ```go http.Handle("/", http.FileServer(imageServer.FileSystem())) go func() { server := &http.Server{ Addr: imagesBindAddr, // default ":8084" ReadHeaderTimeout: 5 * time.Second, } err := server.ListenAndServe() ``` **pkg/ignition/service_config.go:94-100 — pull secret is embedded in every served image** ```go func (b *ignitionBuilder) authFile() ignition_config_types_32.File { source := "data:;base64," + strings.TrimSpace(b.ironicAgentPullSecret) return ignition_config_types_32.File{ Node: ignition_config_types_32.Node{Path: "/etc/authfile.json"}, ... ``` ### Attack Pattern Adjacent provisioning-network client → GET / lists UUID image names → download ISO/initramfs → extract embedded Ignition → recover cluster pull secret and SSH key.
-
CVE-2026-120684
A memory-initialization flaw in the legacy GridFS file handling of the MongoDB C Driver can cause uninitialized process memory to be treated as valid file content and stored on the server. A database user holding ordinary write access to a shared storage area can arrange conditions under which an application later appends to such a file and writes that memory into records the same user is permitted to read, which may result in unintended disclosure of sensitive information from the application's process, including credentials.
-
CVE-2026-120956
The MongoDB C Driver does not validate the type or format of a namespace field returned by the server in a query response. A MongoDB endpoint that a client application connects to, without needing any credentials, can return a malformed value in that field. This causes the client application process to terminate, and can also cause the driver to read memory outside an allocated buffer and to send a small amount of the application's own uninitialized heap memory back to the endpoint in a follow-up command. The impact is limited to the process that embeds the driver.
-
CVE-2026-119922
Improper access control in the Babelfish for PostgreSQL extension versions 1.2-1.5, 2.1-2.15, 3.1-3.12, 4.0-4.8, 5.1-5.4 and in Amazon Aurora PostgreSQL with Babelfish may allow remote authenticated users to elevate their privileges. To remediate this issue, we recommend users upgrade to Babelfish for PostgreSQL extension versions 4.10, 5.6, 6.1 or later. Customers using Amazon Aurora PostgreSQL with Babelfish should upgrade their cluster to Aurora PostgreSQL 13.9.16 (1.5.5), 13.23.3 (1.17.2), 14.6.16 (2.3.5), 14.17.8 (2.12.4), 14.18.6 (2.13.3), 14.19.5 (2.14.2), 14.20.5 (2.15.2), 14.22.3 (2.16.1), 14.23.1 (2.17.0), 15.10.8 (3.8.4), 15.12.8 (3.9.4), 15.13.6 (3.10.3), 15.14.5 (3.11.2), 15.15.5 (3.12.2), 15.17.3 (3.13.1), 15.18.1 (3.14.0), 16.8.8 (4.5.5), 16.9.6 (4.6.3), 16.10.5 (4.7.2), 16.11.5 (4.8.2), 16.13.3 (4.9.1), 16.14.1 (4.10.0), 17.4.8 (5.1.4), 17.5.6 (5.2.3), 17.6.5 (5.3.2), 17.7.5 (5.4.2), 17.9.3 (5.5.1), 17.10.1 (5.6.0), 18.3.4 (6.0.2), 18.4.1 (6.1.0) or later, and apply any pending maintenance to ensure they are on the latest patch.
-
CVE-2026-121587
Coolify before 4.1.1 contains an information disclosure vulnerability that allows unauthenticated attackers to enumerate application names across all teams via manual GitHub, GitLab, Bitbucket and Gitea webhooks. Attackers can send a repository name of % or empty with a common branch like main, exploiting unescaped LIKE wildcards to receive every matching application name before signature verification.
-
CVE-2026-121602
Solanum before commit 2ce64de contains a heap-based buffer overflow vulnerability that allows a malicious linked server to corrupt memory by sending overlong channel names in JOIN or SJOIN messages. Attackers controlling a peer server can trigger strcpy and sprintf overflows in ms_join and ms_sjoin to crash the ircd or potentially achieve remote code execution.
-
CVE-2026-121603
Solanum before commit 2ea1382 contains a buffer overflow vulnerability in ms_kill() that allows malicious linked servers to corrupt memory by sending an overlong kill reason. Attackers controlling a rogue or compromised peer server can send a reason exceeding KILLLEN that overflows a static 512-byte buffer, crashing the ircd or potentially achieving remote code execution.
-
CVE-2026-119675
MongoDB Database Tools mongorestore inconsistently handles the system.buckets. prefix when validating collection names in an untrusted dump. A specially crafted collection name can bypass special-collection guards and operator namespace filters, causing a restore to create an unauthorized user-chosen database account, including an account with the root role. Exploitation requires an operator to restore unauthorized user-controlled data using credentials with sufficient user-management privileges on the destination deployment. Excluding the admin namespace alone does not prevent this behavior. This issue affects MongoDB Database Tools versions 100.4.0 through versions before 100.20.1.
-
CVE-2026-119672
MongoDB Database Tools mongofiles uses a database-stored GridFS filename as the output destination for get_id when --local is not specified. A user who can modify the file metadata can set the stored filename to '-', causing a later authorized retrieval to write the file contents to standard output instead of a local file. If standard output is captured in a location accessible to that user, this may disclose GridFS content the user cannot read directly. Exploitation requires a higher-privileged operator to retrieve the affected file without an explicit local destination and attacker access to the captured output. This issue affects MongoDB Database Tools from version 3.1.4 through versions before 100.20.1.
-
CVE-2026-119673
MongoDB Database Tools mongofiles does not resolve existing parent-directory symlinks when validating output paths for get_regex. An unauthorized user who can both upload a GridFS file and place a symlink in the operator's download working directory can cause a later retrieval to create a file containing unauthorized user-controlled data outside that directory, using the operator process's filesystem permissions. The issue can occur with default traversal and overwrite settings. Exploitation requires an operator to retrieve matching files from the affected working directory; the demonstrated default behavior creates a new file rather than overwriting an existing file. This issue affects MongoDB Database Tools versions 100.2.0 through versions before 100.20.1.
-
CVE-2026-119674
MongoDB Database Tools mongorestore incorrectly URL-decodes an original collection name read from metadata when the corresponding dump filename has been truncated. A database user who can create and populate a suitably encoded collection can cause a later authorized directory restore to write those documents into a different collection that the user cannot modify directly. When both collection names resolve to the same destination, backed-up documents may be omitted; use of --drop can also replace the unintended destination collection. Exploitation requires an operator to restore the affected collection from a directory dump with write access to the destination. This issue affects MongoDB Database Tools versions 100.0.2 through versions before 100.20.1.
-
CVE-2026-115714
Improper Certificate Validation (CWE-295) in the operating system certificate bundle shipped in the Docker images of APM Server; Auditbeat, Filebeat, Heartbeat, Metricbeat and Packetbeat; Elastic Agent, including Fleet Server and Cloudbeat; Logstash, for TLS connections that use the operating system certificate bundle; Elastic connectors; Universal Profiling (host agent, collector and symbolizer); and Elastic Maps Server. This certificate bundle, a dependency of these products (Dependency on Vulnerable Third-Party Component, CWE-1395), could allow an attacker to intercept or modify TLS-protected traffic via Signature Spoofing by Key Theft (CAPEC-474). The certificate bundle includes root certificate authorities that signed eight eMudhra intermediate certificate authorities whose private keys are compromised. Anyone with these keys can issue a certificate for any host name that these products accept as trusted. Exploitation requires an attacker in a network position between the product and the remote party, who presents a certificate issued with one of the compromised keys.
-
CVE-2026-115716
Allocation of Resources Without Limits or Throttling (CWE-770) in APM Server can lead to denial of service via Oversized Serialized Data Payloads (CAPEC-231). An attacker who can send OpenTelemetry (OTLP) data to APM Server over HTTP can cause APM Server to exhaust its memory and terminate. This stops the intake of APM and OpenTelemetry data until APM Server restarts. The issue is resolved in version 8.19.24, 9.4.9 and 9.5.6. The fixed versions reject OTLP/HTTP requests with a body larger than 64 MiB after decompression, as recommended by the OTLP specification.
-
CVE-2026-115298
Incorrect Authorization (CWE-863) in Elasticsearch can lead to information disclosure via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user whose index access is restricted by field-level security can retrieve the value of a constant_keyword field that their role excludes. Some search and query features do not apply the field-level security restriction to this field type.
-
CVE-2026-115710
Use of Cache Containing Sensitive Information (CWE-524) in Elasticsearch can lead to information disclosure and incorrect search results via specially crafted search requests. For some query types, Elasticsearch can share cached query results between users who have field-level security restrictions and users who do not. A user restricted by field-level security can learn which documents match a value in a field that they cannot read. A restricted user can also cause users with full access to receive incomplete search results. Disclosure is only possible when a user with full access ran the same query on the same shard recently.
-
CVE-2026-115289
Use of Multiple Resources with Duplicate Identifier (CWE-694) in APM Server can lead to information disclosure via a specially crafted agent configuration request. An authenticated APM agent could receive the central agent configuration that belongs to a different service until the cached configuration expires. A service could also run with central configuration settings, such as sampling settings, that were meant for a different service.
-
CVE-2026-115290
Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via Serialized Data with Nested Payloads (CAPEC-230). An unauthenticated attacker with network access to the remote cluster server interface of a node can send a specially crafted request that causes the node to terminate. No credentials are required. The attacker can repeat the request against each reachable node.
-
CVE-2026-103916
Docker Sandboxes sbx cp does not safely confine host-side paths to a shared workspace. A malicious sandbox can create or replace a workspace symlink so that a later user-initiated copy reads a host file outside the workspace into the sandbox or writes sandbox-controlled data to a host path outside the workspace with the invoking user's permissions.
-
CVE-2026-65165
NVIDIA Slurm contains a vulnerability where a user could deliberately alter the provided nodelist alongside the --distribution=arbitrary option to gain access to compute nodes outside of the current job allocation.
-
CVE-2026-65168
NVIDIA Slurm contains an attacker-controlled heap-based buffer overflow in the slurmstepd socket handling code.
-
CVE-2026-65108
NVIDIA Slurm contains an attacker-controlled heap-based buffer overflow in the slurmstepd process while the slurmstepd is running with root permissions.
-
CVE-2026-65109
NVIDIA Slurm is subject to a symlink resolution issue in the container cleanup code that could allow an attacker to remove files outside of their control on the compute node.
-
CVE-2026-65139
NVIDIA Slurm contains an attacker-controlled SQL injection attack in certain SlurmDBD RPCs that can lead to arbitrary code execution.
The newest published CVE records from CVE.org and NVD, refreshed every few minutes, with severity, exploit odds and fix status.