{"id":"CVE-2019-10099","published":"2019-08-07T16:18:46Z","modified":"2024-08-04T22:10:09Z","title":"Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if…","vendor":"Apache","product":"Apache Spark","tier":"tail","description":{"text":"Prior to Spark 2.3.3, in certain situations Spark would write user data to local disk unencrypted, even if spark.io.encryption.enabled=true. This includes cached blocks that are fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem); in SparkR, using parallelize; in Pyspark, using broadcast and parallelize; and use of python udfs.","provenance":"source","from":"CVE.org"},"cvss":null,"epss":{"score":0.01308,"percentile":0.71308,"provenance":"source"},"cwes":[],"references":[{"url":"https://lists.apache.org/thread.html/c2a39c207421797f82823a8aff488dcd332d9544038307bf69a2ba9e%40%3Cuser.spark.apache.org%3E","tags":["x_refsource_MISC"]},{"url":"https://lists.apache.org/thread.html/rabe1d47e2bf8b8f6d9f3068c8d2679731d57fa73b3a7ed1fa82406d2%40%3Cissues.spark.apache.org%3E","tags":["mailing-list","x_refsource_MLIST"]},{"url":"https://lists.apache.org/thread.html/ra216b7b0dd82a2c12c2df9d6095e689eb3f3d28164e6b6587da69fae%40%3Ccommits.spark.apache.org%3E","tags":["mailing-list","x_refsource_MLIST"]}],"priority":{"score":14.3,"factors":[{"name":"EPSS percentile","points":14.26}]},"vendorSlug":"apache","affectedVersions":[{"vendor":"Apache","product":"Apache Spark","versions":["2.3.2 and below"]}],"affected":[{"ecosystem":"Maven","package":"org.apache.spark:spark-core_2.11","vulnerableRanges":["<2.3.3"],"patched":["2.3.3"],"from":["GHSA-fp5j-3fpf-mhj5"],"provenance":"source"},{"ecosystem":"PyPI","package":"pyspark","vulnerableRanges":["<2.3.3"],"patched":["2.3.3"],"from":["GHSA-fp5j-3fpf-mhj5","PYSEC-2019-114"],"provenance":"source"}],"summary":{"text":"Apache Spark prior to 2.3.3 has a flaw in its local disk handling that can leave user data unencrypted on disk, even when spark.io.encryption.enabled=true is set. The affected paths include cached blocks fetched to disk (controlled by spark.maxRemoteBlockSizeFetchToMem), parallelize in SparkR, broadcast and parallelize in PySpark, and Python UDFs.","provenance":"ai","model":"claude-haiku-5-5","generatedAt":"2026-11-20T21:49:57Z"},"euvd":{"id":"EUVD-2019-0110","exploited":false,"provenance":"source"},"lineage":{"packageHistory":[{"cve":"CVE-2017-7678","published":"2017-07-12","fixed":"2.2.0","kev":false},{"cve":"CVE-2018-1334","published":"2018-07-12","fixed":"2.3.1","kev":false},{"cve":"CVE-2018-8024","published":"2018-07-12","fixed":"2.3.1","kev":false,"link":{"from":"CVE-2018-8024","to":"CVE-2019-10099","relation":"successor","confidence":"low","kind":"sequence","evidence":[{"kind":"version-step","provenance":"source","package":"Maven:org.apache.spark:spark-core_2.11","fromFixed":"2.3.1","toFixed":"2.3.3","daysApart":391}]}},{"cve":"CVE-2019-10099","published":"2019-08-07","fixed":"2.3.3","kev":false},{"cve":"CVE-2022-31777","published":"2022-11-01","fixed":"3.2.2","kev":false}],"package":"Maven:org.apache.spark:spark-core_2.11"}}