CVE record
CVE-2019-10099
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.
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.
CVE evidence
Known exploitation
Not listed in fetched KEV catalog
Absence from the fetched catalog does not establish that exploitation has not occurred.
Severity
high
CVSS 8.7 · CVSS_V4
Affected packages
2
Supported package records returned by OSV. Vendor and product names are not used to infer matches.
OSV package mapping
Affected open-source packages
| Package | Ecosystem | Fixed versions |
|---|---|---|
| org.apache.spark:spark-core_2.11 | maven | 2.3.3 |
| pyspark | pypi | 2.3.3 |