Skip to content
Heist/ feed
Back to feed
High

Apache Airflow: Unguarded import_string() of airflow_exc_ser / base_exc_ser exception nodes in BaseSerialization.deserialize enables DAG-author RCE on Scheduler / API Server

Type: advisorySources: OSV

Apache Airflow's serialization layer reconstructed exception nodes by calling `import_string()` on a class name taken from the serialized blob and instantiating it with arguments from the same blob, with no restriction on what could be imported. An operator's `executor_config` reaches that branch, so a Dag author could place a value there that causes an arbitrary callable to be imported and invoked -- for example `subprocess.check_output`, or `builtins.eval` on the `builtins`-prefixed variant. The code runs in the **Scheduler**, which reconstructs serialized Dags in its normal loop with no request involved, and in the **API server**, on any authenticated read of the Dag such as `GET /api/v2/dags/{dag_id}/details`. Both are components the Airflow security model states must never execute Dag-author code, and both hold the metadata database credentials and the JWT signing secret. No non-default configuration is required. This is a **different sink from CVE-2026-33264**, which covered only the trigger branch of the same deserializer: deployments that upgraded in response to that advisory are still affected through the exception branch and must upgrade again. Users are advised to upgrade to apache-airflow 3.3.1 or later, which restricts the imported class to a subclass of `BaseException`.

Products
Not specified
Affected versions
Not specified
Fixed versions
Not specified
CVEs
CVE-2026-58076
Confirmed fixed CVEs
Not specified
Updated in this feed
8 Nov 2026, 01:47 UTC · Revision 1

Source evidence

Apache Airflow: Unguarded import_string() of airflow_exc_ser / base_exc_ser exception nodes in BaseSerialization.deserialize enables DAG-author RCE on Scheduler / API Server

core-osv · Original source

Published 12 Aug 2026, 15:40 UTC

Apache Airflow's serialization layer reconstructed exception nodes by calling `import_string()` on a class name taken from the serialized blob and instantiating it with arguments from the same blob, with no restriction on what could be imported. An operator's `executor_config` reaches that branch, so a Dag author could place a value there that causes an arbitrary callable to be imported and invoked -- for example `subprocess.check_output`, or `builtins.eval` on the `builtins`-prefixed variant. The code runs in the **Scheduler**, which reconstructs serialized Dags in its normal loop with no request involved, and in the **API server**, on any authenticated read of the Dag such as `GET /api/v2/dags/{dag_id}/details`. Both are components the Airflow security model states must never execute Dag-author code, and both hold the metadata database credentials and the JWT signing secret. No non-default configuration is required. This is a **different sink from CVE-2026-33264**, which covered only the trigger branch of the same deserializer: deployments that upgraded in response to that advisory are still affected through the exception branch and must upgrade again. Users are advised to upgrade to apache-airflow 3.3.1 or later, which restricts the imported class to a subclass of `BaseException`.

osv:CVE-2026-58076
Complete source record
{
  "id": "2728f0a23fecea077581d8666137e9a8cdaaf0efb6386627f71d8a1a7eebcd24",
  "url": "https://osv.dev/vulnerability/CVE-2026-58076",
  "method": "structured",
  "payload": {
    "kind": "advisory",
    "title": "Apache Airflow: Unguarded import_string() of airflow_exc_ser / base_exc_ser exception nodes in BaseSerialization.deserialize enables DAG-author RCE on Scheduler / API Server",
    "summary": "Apache Airflow's serialization layer reconstructed exception nodes by calling `import_string()` on a class name taken from the serialized blob and instantiating it with arguments from the same blob, with no restriction on what could be imported. An operator's `executor_config` reaches that branch, so a Dag author could place a value there that causes an arbitrary callable to be imported and invoked -- for example `subprocess.check_output`, or `builtins.eval` on the `builtins`-prefixed variant. The code runs in the **Scheduler**, which reconstructs serialized Dags in its normal loop with no request involved, and in the **API server**, on any authenticated read of the Dag such as `GET /api/v2/dags/{dag_id}/details`. Both are components the Airflow security model states must never execute Dag-author code, and both hold the metadata database credentials and the JWT signing secret. No non-default configuration is required. This is a **different sink from CVE-2026-33264**, which covered only the trigger branch of the same deserializer: deployments that upgraded in response to that advisory are still affected through the exception branch and must upgrade again. Users are advised to upgrade to apache-airflow 3.3.1 or later, which restricts the imported class to a subclass of `BaseException`.",
    "evidence": "osv:CVE-2026-58076",
    "products": [],
    "severity": "high",
    "structured": {
      "id": "CVE-2026-58076",
      "aliases": [
        "BIT-airflow-2026-58076",
        "CVE-2026-58076",
        "GHSA-8prw-8m6x-8w6f",
        "PYSEC-2026-3991"
      ],
      "metrics": [
        {
          "type": "CVSS_V3",
          "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H"
        }
      ],
      "affected": [
        {
          "ranges": [
            {
              "repo": "https://github.com/apache/airflow",
              "type": "GIT",
              "events": [
                {
                  "introduced": "1fc3442f1307fdf5e1221b26f41266c4fed330d3"
                },
                {
                  "fixed": "275ac7303fa702e232c277ac7f4929baa85464d2"
                }
              ],
              "database_specific": {
                "extracted_events": [
                  {
                    "range": [
                      {
                        "introduced": "3.0.0"
                      },
                      {
                        "fixed": "3.3.1"
                      }
                    ],
                    "source": "AFFECTED_FIELD",
                    "strategy": "StandardRange"
                  },
                  {
                    "cpe": "cpe:2.3:a:apache:airflow:*:*:*:*:*:*:*:*",
                    "range": [
                      {
                        "introduced": "3.3.0"
                      },
                      {
                        "fixed": "3.3.1"
                      }
                    ],
                    "source": "CPE_RANGE"
                  }
                ]
              }
            }
          ],
          "database_specific": {
            "source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-58076.json"
          }
        }
      ],
      "modified": "2026-10-28T02:30:52.740894249Z",
      "provider": "osv",
      "references": [
        "https://pypi.python.org",
        "https://www.cve.org/CVERecord?id=CVE-2026-33264",
        "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/58xxx/CVE-2026-58076.json",
        "https://lists.apache.org/thread/t81p688t15jozxsng8521o60nh2kfsos",
        "https://nvd.nist.gov/vuln/detail/CVE-2026-58076",
        "https://github.com/apache/airflow/pull/68511"
      ],
      "provider_metadata": {
        "cwe_ids": [
          "CWE-502"
        ],
        "cna_assigner": "apache",
        "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/58xxx/CVE-2026-58076.json"
      }
    },
    "related_cves": [
      {
        "value": "CVE-2026-58076",
        "evidence": "/id"
      }
    ],
    "security_relevant": true,
    "severity_evidence": "/severity/0/score"
  },
  "source_id": "core-osv",
  "document_id": 521487,
  "published_at": "2026-08-12T15:40:59.934+00:00"
}