The Keycloak authentication manager in `apache-airflow-providers-keycloak` did not generate or validate the OAuth 2.0 `state` parameter on the login / login-callback flow, and did not use PKCE. An attacker with a Keycloak account in the same realm could deliver a crafted callback URL to a victim's browser and cause the victim to be logged into the attacker's Airflow session (login-CSRF / session fixation), where any credentials the victim subsequently stored in Airflow Connections would be harvestable by the attacker. Users are advised to upgrade `apache-airflow-providers-keycloak` to 0.7.0 or later.

Secrets in Variables saved as JSON dictionaries were not properly redacted - in case thee variables were retrieved by the user the secrets stored as nested fields were not masked. If you do not store variables with sensitive values in JSON form, you are not affected. Otherwise please upgrade to Apache Airflow 3.2.0 that has the fix implemented

An example of BashOperator in Airflow documentation suggested a way of passing dag_run.conf in the way that could cause unsanitized user input to be used to escalate privileges of UI user to allow execute code on worker. Users should review if any of their own DAGs have adopted this incorrect advice.

In case of SQL errors, exception/stack trace of errors was exposed in API even if "api/expose_stack_traces" was set to false. That could lead to exposing additional information to potential attacker. Users are recommended to upgrade to Apache Airflow 3.2.0, which fixes the issue.

Dag Authors, who normally should not be able to execute code in the webserver context could craft XCom payload causing the webserver to execute arbitrary code. Since Dag Authors are already highly trusted, severity of this issue is Low. Users are recommended to upgrade to Apache Airflow 3.2.0, which fixes the issue.

UI / API User with asset materialize permission could trigger dags they had no access to. Users are advised to migrate to Airflow version 3.2.0 that fixes the issue.

JWT Tokens used by tasks were exposed in logs. This could allow UI users to act as Dag Authors. Users are advised to upgrade to Airflow version that contains fix. Users are recommended to upgrade to version 3.2.0, which fixes this issue.

The `access_key` and `connection_string` connection properties were not marked as sensitive names in secrets masker. This means that user with read permission could see the values in Connection UI, as well as when Connection was accidentaly logged to logs, those values could be seen in the logs. Azure Service Bus used those properties to store sensitive values. Possibly other providers could be also affected if they used the same fields to store sensitive data. If you used Azure Service Bus connection with those values set or if you have other connections with those values storing sensitve values, you should upgrade Airflow to 3.1.8

The example example_xcom that was included in airflow documentation implemented unsafe pattern of reading value from xcom in the way that could be exploited to allow UI user who had access to modify XComs to perform arbitrary execution of code on the worker. Since the UI users are already highly trusted, this is a Low severity vulnerability. It does not affect Airflow release - example_dags are not supposed to be enabled in production environment, however users following the example could replicate the bad pattern. Documentation of Airflow 3.2.0 contains version of the example with improved resiliance for that case. Users who followed that pattern are advised to adjust their implementations accordingly.

Dag Authors, who normally should not be able to execute code in the webserver context could craft XCom payload causing the webserver to execute arbitrary code. Since Dag Authors are already highly trusted, severity of this issue is Low. Users are recommended to upgrade to Apache Airflow 3.2.0, which resolves this issue.

Before Airflow 3.2.0, it was unclear that secure Airflow deployments require the Deployment Manager to take appropriate actions and pay attention to security details and security model of Airflow. Some assumptions the Deployment Manager could make were not clear or explicit enough, even though Airflow's intentions and security model of Airflow did not suggest different assumptions. The overall security model [1], workload isolation [2], and JWT authentication details [3] are now described in more detail. Users concerned with role isolation and following the Airflow security model of Airflow are advised to upgrade to Airflow 3.2, where several security improvements have been implemented. They should also read and follow the relevant documents to make sure that their deployment is secure enough. It also clarifies that the Deployment Manager is ultimately responsible for securing your Airflow deployment. This had also been communicated via Airflow 3.2.0 Blog announcement [4]. [1] Security Model: https://airflow.apache.org/docs/apache-airflow/stable/security/jwt_token_authentication.html [2] Workload isolation: https://airflow.apache.org/docs/apache-airflow/stable/security/workload.html [3] JWT Token authentication: https://airflow.apache.org/docs/apache-airflow/stable/security/jwt_token_authentication.html [4] Airflow 3.2.0 Blog announcement: https://airflow.apache.org/blog/airflow-3.2.0/ Users are recommended to upgrade to version 3.2.0, which fixes this issue.

When user logged out, the JWT token the user had authtenticated with was not invalidated, which could lead to reuse of that token in case it was intercepted. In Airflow 3.2 we implemented the mechanism that implements token invalidation at logout. Users who are concerned about the logout scenario and possibility of intercepting the tokens, should upgrade to Airflow 3.2+ Users are recommended to upgrade to version 3.2.0, which fixes this issue.

Apache Airflow versions 3.0.0 through 3.1.8 DagRun wait endpoint returns XCom result values even to users who only have DAG Run read permissions, such as the Viewer role.This behavior conflicts with the FAB RBAC model, which treats XCom as a separate protected resource, and with the security model documentation that defines the Viewer role as read-only. Airflow uses the FAB Auth Manager to manage access control on a per-resource basis. The Viewer role is intended to be read-only by default, and the security model documentation defines Viewer users as those who can inspect DAGs without accessing sensitive execution results. Users are recommended to upgrade to Apache Airflow 3.2.0 which resolves this issue.

Apache Airflow versions 3.1.0 through 3.1.7 /ui/dependencies endpoint returns the full DAG dependency graph without filtering by authorized DAG IDs. This allows an authenticated user with only DAG Dependencies permission to enumerate DAGs they are not authorized to view. Users are recommended to upgrade to Apache Airflow 3.1.8 or later, which resolves this issue.

Apache Airflow versions 3.0.0 through 3.1.7 FastAPI DagVersion listing API does not apply per-DAG authorization filtering when the request is made with dag_id set to "~" (wildcard for all DAGs). As a result, version metadata of DAGs that the requester is not authorized to access is returned. Users are recommended to upgrade to Apache Airflow 3.1.8 or later, which resolves this issue.

Apache Airflow versions 3.1.0 through 3.1.7 missing authorization vulnerability in the Execution API's Human-in-the-Loop (HITL) endpoints that allows any authenticated task instance to read, approve, or reject HITL workflows belonging to any other task instance. Users are recommended to upgrade to Apache Airflow 3.1.8 or later, which resolves this issue.

Apache Airflow versions 3.1.0 through 3.1.7 session token (_token) in cookies is set to path=/ regardless of the configured [webserver] base_url or [api] base_url. This allows any application co-hosted under the same domain to capture valid Airflow session tokens from HTTP request headers, allowing full session takeover without attacking Airflow itself. Users are recommended to upgrade to Apache Airflow 3.1.8 or later, which resolves this issue.

Airflow versions before 2.11.1 have a vulnerability that allows authenticated users with audit log access to see sensitive values in audit logs which they should not see. When sensitive connection parameters were set via airflow CLI, values of those variables appeared in the audit log and were stored unencrypted in the Airflow database. While this risk is limited to users with audit log access, it is recommended to upgrade to Airflow 2.11.1 or a later version, which addresses this issue. Users who previously used the CLI to set connections should manually delete entries with those connection sensitive values from the log table. This is similar but not the same issue as CVE-2024-50378

DAG Author (who already has quite a lot of permissions) could manipulate database of Airflow 2 in the way to execute arbitrary code in the web-server context, which they should normally not be able to do, leading to potentially remote code execution in the context of web-server (server-side) as a result of a user viewing historical task information. The functionality responsible for that (log template history) has been disabled by default in 2.11.1 and users should upgrade to Airflow 3 if they want to continue to use log template history. They can also manually modify historical log file names if they want to see historical logs that were generated before the last log template change.

When a DAG failed during parsing, Airflow’s error-reporting in the UI could include the full kwargs passed to the operators. If those kwargs contained sensitive values (such as secrets), they might be exposed in the UI tracebacks to authenticated users who had permission to view that DAG.  The issue has been fixed in Airflow 3.1.4 and 2.11.1, and users are strongly advised to upgrade to prevent potential disclosure of sensitive information.

Apache Airflow versions 3.1.0 through 3.1.6 contain an authorization flaw that can allow an authenticated user with custom permissions limited to task access to view task logs without having task log access. Users are recommended to upgrade to Apache Airflow 3.1.7 or later, which resolves this issue.

Apache Airflow versions 3.0.0 - 3.1.7, has vulnerability that allows authenticated UI users with permission to one or more specific Dags to view import errors generated by other Dags they did not have access to. Users are advised to upgrade to 3.1.7 or later, which resolves this issue

In Apache Airflow versions before 3.1.6, and 2.11.1 the proxies and proxy fields within a Connection may include proxy URLs containing embedded authentication information. These fields were not treated as sensitive by default and therefore were not automatically masked in log output. As a result, when such connections are rendered or printed to logs, proxy credentials embedded in these fields could be exposed. Users are recommended to upgrade to 3.1.6 or later for Airflow 3, and 2.11.1 or later for Airflow 2 which fixes this issue

In Apache Airflow versions before 3.1.6, when rendered template fields in a Dag exceed [core] max_templated_field_length, sensitive values could be exposed in cleartext in the Rendered Templates UI. This occurred because serialization of those fields used a secrets masker instance that did not include user-registered mask_secret() patterns, so secrets were not reliably masked before truncation and display. Users are recommended to upgrade to 3.1.6 or later, which fixes this issue

A vulnerability in Apache Airflow allowed authenticated UI users to view secret values in rendered templates due to secrets not being properly redacted, potentially exposing secrets to users without the appropriate authorization. Users are recommended to upgrade to version 3.1.4, which fixes this issue.

An example dag `example_dag_decorator` had non-validated parameter that allowed the UI user to redirect the example to a malicious server and execute code on worker. This however required that the example dags are enabled in production (not default) or the example dag code copied to build your own similar dag. If you used the `example_dag_decorator` please review it and apply the changes implemented in Airflow 3.0.5 accordingly.

API users via `/api/v2/dagReports` could perform Dag code execution in the context of the api-server if the api-server was deployed in the environment where Dag files were available.

User with CREATE and no UPDATE privilege for Pools, Connections, Variables could update existing records via bulk create API with overwrite action.

Apache Airflow 3 introduced a change to the handling of sensitive information in Connections. The intent was to restrict access to sensitive connection fields to Connection Editing Users, effectively applying a "write-only" model for sensitive values. In Airflow 3.0.3, this model was unintentionally violated: sensitive connection information could be viewed by users with READ permissions through both the API and the UI. This behavior also bypassed the `AIRFLOW__CORE__HIDE_SENSITIVE_VAR_CONN_FIELDS` configuration option. This issue does not affect Airflow 2.x, where exposing sensitive information to connection editors was the intended and documented behavior. Users of Airflow 3.0.3 are advised to upgrade Airflow to >=3.0.4.

Apache Airflow versions before 2.10.3 contain a vulnerability that could expose sensitive configuration variables in task logs. This vulnerability allows DAG authors to unintentionally or intentionally log sensitive configuration variables. Unauthorized users could access these logs, potentially exposing critical data that could be exploited to compromise the security of the Airflow deployment. In version 2.10.3, secrets are now masked in task logs to prevent sensitive configuration variables from being exposed in the logging output. Users should upgrade to Airflow 2.10.3 or the latest version to eliminate this vulnerability. If you suspect that DAG authors could have logged the secret values to the logs and that your logs are not additionally protected, it is also recommended that you update those secrets.

Airflow versions before 2.10.3 have a vulnerability that allows authenticated users with audit log access to see sensitive values in audit logs which they should not see. When sensitive variables were set via airflow CLI, values of those variables appeared in the audit log and were stored unencrypted in the Airflow database. While this risk is limited to users with audit log access, it is recommended to upgrade to Airflow 2.10.3 or a later version, which addresses this issue. Users who previously used the CLI to set secret variables should manually delete entries with those variables from the log table.

Apache Airflow versions before 2.10.1 have a vulnerability that allows DAG authors to add local settings to the DAG folder and get it executed by the scheduler, where the scheduler is not supposed to execute code submitted by the DAG author. Users are advised to upgrade to version 2.10.1 or later, which has fixed the vulnerability.

Example DAG: example_inlet_event_extra.py shipped with Apache Airflow version 2.10.0 has a vulnerability that allows an authenticated attacker with only DAG trigger permission to execute arbitrary commands. If you used that example as the base of your DAGs - please review if you have not copied the dangerous example; see https://github.com/apache/airflow/pull/41873  for more information. We recommend against exposing the example DAGs in your deployment. If you must expose the example DAGs, upgrade Airflow to version 2.10.1 or later.

Apache Airflow, versions before 2.10.0, have a vulnerability that allows the developer of a malicious provider to execute a cross-site scripting attack when clicking on a provider documentation link. This would require the provider to be installed on the web server and the user to click the provider link. Users should upgrade to 2.10.0 or later, which fixes this vulnerability.

Apache Airflow 2.4.0, and versions before 2.9.3, has a vulnerability that allows authenticated DAG authors to craft a doc_md parameter in a way that could execute arbitrary code in the scheduler context, which should be forbidden according to the Airflow Security model. Users should upgrade to version 2.9.3 or later which has removed the vulnerability.

Apache Airflow versions before 2.9.3 have a vulnerability that allows an authenticated attacker to inject a malicious link when installing a provider. Users are recommended to upgrade to version 2.9.3, which fixes this issue.

Use of Web Browser Cache Containing Sensitive Information vulnerability in Apache Airflow.  Airflow did not return "Cache-Control" header for dynamic content, which in case of some browsers could result in potentially storing sensitive data in local cache of the browser. This issue affects Apache Airflow: before 2.9.2. Users are recommended to upgrade to version 2.9.2, which fixes the issue.

Apache Airflow version 2.9.0 has a vulnerability that allows an authenticated attacker to inject malicious data into the task instance logs.  Users are recommended to upgrade to version 2.9.1, which fixes this issue.

Airflow versions 2.7.0 through 2.8.4 have a vulnerability that allows an authenticated user to see sensitive provider configuration via the "configuration" UI page when "non-sensitive-only" was set as "webserver.expose_config" configuration (The celery provider is the only community provider currently that has sensitive configurations). You should migrate to Airflow 2.9 or change your "expose_config" configuration to False as a workaround. This is similar, but different to CVE-2023-46288 https://github.com/advisories/GHSA-9qqg-mh7c-chfq which concerned API, not UI configuration page.

Improper Preservation of Permissions vulnerability in Apache Airflow.This issue affects Apache Airflow from 2.8.2 through 2.8.3. Airflow's local file task handler in Airflow incorrectly set permissions for all parent folders of log folder, in default configuration adding write access to Unix group of the folders. In the case Airflow is run with the root user (not recommended) it added group write permission to all folders up to the root of the filesystem. If your log files are stored in the home directory, these permission changes might impact your ability to run SSH operations after your home directory becomes group-writeable. This issue does not affect users who use or extend Airflow using Official Airflow Docker reference images ( https://hub.docker.com/r/apache/airflow/ ) - those images require to have group write permission set anyway. You are affected only if you install Airflow using local installation / virtualenv or other Docker images, but the issue has no impact if docker containers are used as intended, i.e. where Airflow components do not share containers with other applications and users. Also you should not be affected if your umask is 002 (group write enabled) - this is the default on many linux systems. Recommendation for users using Airflow outside of the containers: * if you are using root to run Airflow, change your Airflow user to use non-root * upgrade Apache Airflow to 2.8.4 or above * If you prefer not to upgrade, you can change the https://airflow.apache.org/docs/apache-airflow/stable/configurations-ref.html#file-task-handler-new-folder-permissions  to 0o755 (original value 0o775). * if you already ran Airflow tasks before and your default umask is 022 (group write disabled) you should stop Airflow components, check permissions of AIRFLOW_HOME/logs in all your components and all parent directories of this directory and remove group write access for all the parent directories

Apache Airflow, versions 2.8.0 through 2.8.2, has a vulnerability that allows an authenticated user with limited permissions to access resources such as variables, connections, etc from the UI which they do not have permission to access.  Users of Apache Airflow are recommended to upgrade to version 2.8.3 or newer to mitigate the risk associated with this vulnerability

Apache Airflow, versions before 2.8.2, has a vulnerability that allows authenticated Ops and Viewers users to view all information on audit logs, including dag names and usernames they were not permitted to view. With 2.8.2 and newer, Ops and Viewer users do not have audit log permission by default, they need to be explicitly granted permissions to see the logs. Only admin users have audit log permission by default. Users of Apache Airflow are recommended to upgrade to version 2.8.2 or newer to mitigate the risk associated with this vulnerability

Apache Airflow, versions before 2.8.2, has a vulnerability that allows authenticated users to view DAG code and import errors of DAGs they do not have permission to view through the API and the UI. Users of Apache Airflow are recommended to upgrade to version 2.8.2 or newer to mitigate the risk associated with this vulnerability

Apache Airflow, versions before 2.8.1, have a vulnerability that allows an authenticated user to access the source code of a DAG to which they don't have access. This vulnerability is considered low since it requires an authenticated user to exploit it. Users are recommended to upgrade to version 2.8.1, which fixes this issue.

Apache Airflow, versions before 2.8.1, have a vulnerability that allows a potential attacker to poison the XCom data by bypassing the protection of "enable_xcom_pickling=False" configuration setting resulting in poisoned data after XCom deserialization. This vulnerability is considered low since it requires a DAG author to exploit it. Users are recommended to upgrade to version 2.8.1 or later, which fixes this issue.

Since version 5.2.0, when using deferrable mode with the path of a Kubernetes configuration file for authentication, the Airflow worker serializes this configuration file as a dictionary and sends it to the triggerer by storing it in metadata without any encryption. Additionally, if used with an Airflow version between 2.3.0 and 2.6.0, the configuration dictionary will be logged as plain text in the triggerer service without masking. This allows anyone with access to the metadata or triggerer log to obtain the configuration file and use it to access the Kubernetes cluster. This behavior was changed in version 7.0.0, which stopped serializing the file contents and started providing the file path instead to read the contents into the trigger. Users are recommended to upgrade to version 7.0.0, which fixes this issue.

Apache Airflow, in versions prior to 2.8.0, contains a security vulnerability that allows an authenticated user with limited access to some DAGs, to craft a request that could give the user write access to various DAG resources for DAGs that the user had no access to, thus, enabling the user to clear DAGs they shouldn't. This is a missing fix for CVE-2023-42792 in Apache Airflow 2.7.2  Users of Apache Airflow are strongly advised to upgrade to version 2.8.0 or newer to mitigate the risk associated with this vulnerability.

Apache Airflow, versions before 2.8.0, is affected by a vulnerability that allows an authenticated user without the variable edit permission, to update a variable. This flaw compromises the integrity of variable management, potentially leading to unauthorized data modification. Users are recommended to upgrade to 2.8.0, which fixes this issue

Apache Airflow, versions 2.6.0 through 2.7.3 has a stored XSS vulnerability that allows a DAG author to add an unbounded and not-sanitized javascript in the parameter description field of the DAG. This Javascript can be executed on the client side of any of the user who looks at the tasks in the browser sandbox. While this issue does not allow to exit the browser sandbox or manipulation of the server-side data - more than the DAG author already has, it allows to modify what the user looking at the DAG details sees in the browser - which opens up all kinds of possibilities of misleading other users. Users of Apache Airflow are recommended to upgrade to version 2.8.0 or newer to mitigate the risk associated with this vulnerability

Apache Airflow, version 2.7.0 through 2.7.3, has a vulnerability that allows an attacker to trigger a DAG in a GET request without CSRF validation. As a result, it was possible for a malicious website opened in the same browser - by the user who also had Airflow UI opened - to trigger the execution of DAGs without the user's consent. Users are advised to upgrade to version 2.8.0 or later which is not affected