Improper Authentication in Apache Spark
Apache Software Foundation Apache Spark
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| org.apache.spark:spark-parent_2.11 Maven |
|
|
| pyspark PyPI |
|
|
Apache Software Foundation Apache Spark
| Product | Affected | Fixed in |
|---|---|---|
| org.apache.spark:spark-parent_2.11 Maven |
|
|
| pyspark PyPI |
|
|
In Apache Spark 2.4.5 and earlier, a standalone resource manager's master may be configured to require authentication (spark.authenticate) via a shared secret. When enabled, however, a specially-crafted RPC to the master can succeed in starting an application's resources on the Spark cluster, even without the shared key. This can be leveraged to execute shell commands on the host machine. This does not affect Spark clusters using other resource managers (YARN, Mesos, etc).
From the CVE record · apache
In Apache Spark 2.4.5 and earlier, a standalone master configured to require shared-secret authentication can accept a specially crafted RPC without the shared key. An unauthenticated attacker can use it to start application resources and execute shell commands on the host; clusters using YARN, Mesos, or other resource managers are not affected.
Operators of standalone Apache Spark clusters and users of Maven org.apache.spark:spark-parent_2.11 or PyPI pyspark versions earlier than 2.4.6 should check their deployments.