Apache Spark uses the standard process outlined by the Apache Security Team for reporting vulnerabilities. Note that vulnerabilities should not be publicly disclosed until the project has responded.
To report a possible security vulnerability, please email security@spark.apache.org
. This is a non-public list that will reach the Apache Security team, as well as the Spark PMC.
Severity: Important
Vendor: The Apache Software Foundation
Versions Affected:
Description:
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).
Mitigation:
Credit:
Severity: Important
Vendor: The Apache Software Foundation
Versions affected:
Description:
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.
Mitigation:
Credit:
Severity: Important
Vendor: The Apache Software Foundation
Versions affected:
Description:
When using PySpark, it's possible for a different local user to connect to the Spark application and impersonate the user running the Spark application. This affects versions 1.x, 2.0.x, 2.1.x, 2.2.0 to 2.2.2, and 2.3.0 to 2.3.1.
Mitigation:
Credit:
Severity: Low
Vendor: The Apache Software Foundation
Versions Affected:
Description:
Spark's standalone resource manager accepts code to execute on a ‘master’ host, that then runs that code on ‘worker’ hosts. The master itself does not, by design, execute user code. A specially-crafted request to the master can, however, cause the master to execute code too. Note that this does not affect standalone clusters with authentication enabled. While the master host typically has less outbound access to other resources than a worker, the execution of code on the master is nevertheless unexpected.
Mitigation:
Enable authentication on any Spark standalone cluster that is not otherwise secured from unwanted access, for example by network-level restrictions. Use spark.authenticate
and related security properties described at https://spark.apache.org/docs/latest/security.html
Credit:
Severity: Low
Vendor: The Apache Software Foundation
Versions Affected
Description:
Spark's Apache Maven-based build includes a convenience script, ‘build/mvn’, that downloads and runs a zinc server to speed up compilation. This server will accept connections from external hosts by default. A specially-crafted request to the zinc server could cause it to reveal information in files readable to the developer account running the build. Note that this issue does not affect end users of Spark, only developers building Spark from source code.
Mitigation:
Credit:
Severity: Medium
Vendor: The Apache Software Foundation
Versions Affected:
Description:
From version 1.3.0 onward, Spark's standalone master exposes a REST API for job submission, in addition to the submission mechanism used by spark-submit
. In standalone, the config property spark.authenticate.secret
establishes a shared secret for authenticating requests to submit jobs via spark-submit
. However, the REST API does not use this or any other authentication mechanism, and this is not adequately documented. In this case, a user would be able to run a driver program without authenticating, but not launch executors, using the REST API. This REST API is also used by Mesos, when set up to run in cluster mode (i.e., when also running MesosClusterDispatcher
), for job submission. Future versions of Spark will improve documentation on these points, and prohibit setting spark.authenticate.secret
when running the REST APIs, to make this clear. Future versions will also disable the REST API by default in the standalone master by changing the default value of spark.master.rest.enabled
to false
.
Mitigation:
For standalone masters, disable the REST API by setting spark.master.rest.enabled
to false
if it is unused, and/or ensure that all network access to the REST API (port 6066 by default) is restricted to hosts that are trusted to submit jobs. Mesos users can stop the MesosClusterDispatcher
, though that will prevent them from running jobs in cluster mode. Alternatively, they can ensure access to the MesosRestSubmissionServer
(port 7077 by default) is restricted to trusted hosts.
Credit:
Severity: Medium
Versions Affected:
Description:
In Apache Spark 2.1.0 to 2.1.2, 2.2.0 to 2.2.1, and 2.3.0, it‘s possible for a malicious user to construct a URL pointing to a Spark cluster’s UI‘s job and stage info pages, and if a user can be tricked into accessing the URL, can be used to cause script to execute and expose information from the user’s view of the Spark UI. While some browsers like recent versions of Chrome and Safari are able to block this type of attack, current versions of Firefox (and possibly others) do not.
Mitigation:
Credit:
Severity: High
Vendor: The Apache Software Foundation
Versions affected:
Description:
In Apache Spark up to and including 2.1.2, 2.2.0 to 2.2.1, and 2.3.0, when using PySpark or SparkR, it's possible for a different local user to connect to the Spark application and impersonate the user running the Spark application.
Mitigation:
Credit:
JIRA: SPARK-20922
Severity: Medium
Vendor: The Apache Software Foundation
Versions Affected:
Description:
In Apache Spark 1.6.0 until 2.1.1, the launcher API performs unsafe deserialization of data received by its socket. This makes applications launched programmatically using the launcher API potentially vulnerable to arbitrary code execution by an attacker with access to any user account on the local machine. It does not affect apps run by spark-submit or spark-shell. The attacker would be able to execute code as the user that ran the Spark application. Users are encouraged to update to version 2.1.2, 2.2.0 or later.
Mitigation:
Update to Apache Spark 2.1.2, 2.2.0 or later.
Credit:
JIRA: SPARK-20393
Severity: Medium
Vendor: The Apache Software Foundation
Versions Affected:
Description:
It is possible for an attacker to take advantage of a user's trust in the server to trick them into visiting a link that points to a shared Spark cluster and submits data including MHTML to the Spark master, or history server. This data, which could contain a script, would then be reflected back to the user and could be evaluated and executed by MS Windows-based clients. It is not an attack on Spark itself, but on the user, who may then execute the script inadvertently when viewing elements of the Spark web UIs.
Mitigation:
Update to Apache Spark 2.1.2, 2.2.0 or later.
Example:
Request:
GET /app/?appId=Content-Type:%20multipart/related;%20boundary=_AppScan%0d%0a-- _AppScan%0d%0aContent-Location:foo%0d%0aContent-Transfer- Encoding:base64%0d%0a%0d%0aPGh0bWw%2bPHNjcmlwdD5hbGVydCgiWFNTIik8L3NjcmlwdD48L2h0bWw%2b%0d%0a HTTP/1.1
Excerpt from response:
<div class="row-fluid">No running application with ID Content-Type: multipart/related; boundary=_AppScan --_AppScan Content-Location:foo Content-Transfer-Encoding:base64 PGh0bWw+PHNjcmlwdD5hbGVydCgiWFNTIik8L3NjcmlwdD48L2h0bWw+ </div>
Result: In the above payload the BASE64 data decodes as:
<html><script>alert("XSS")</script></html>
Credit: