<?xml version="1.0" encoding="UTF-8"?>
<!-- AUTOGENERATED FILE. DO NOT EDIT. -->
<feed xmlns="http://www.w3.org/2005/Atom">
  <id>tag:google.com,2016:dataproc-release-notes</id>
  <title>Dataproc - Release notes</title>
  <link rel="self" href="https://docs.cloud.google.com/feeds/dataproc-release-notes.xml"/>
  <author>
    <name>Google Cloud Platform</name>
  </author>
  <updated>2026-04-03T00:00:00-07:00</updated>

  <entry>
    <title>April 03, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#April_03_2026</id>
    <updated>2026-04-03T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#April_03_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p><strong>Dataproc</strong> and <strong>Google Cloud Serverless for Apache Spark</strong> are now unified under the <a href="https://console.cloud.google.com/dataproc/overview"><strong>Managed Service for Apache Spark</strong></a> brand. This change consolidates our managed Spark deployment options into a single umbrella brand that includes the full breadth of our Spark capabilities. No existing functionality is being removed as part of this change, and there will be no impact to the Dataproc API, client library, CLI, or IAM names. </p>
]]>
    </content>
  </entry>

  <entry>
    <title>April 02, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#April_02_2026</id>
    <updated>2026-04-02T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#April_02_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.3.28-debian12, 2.3.28-ml-ubuntu22, 2.3.28-rocky9, 2.3.28-ubuntu22, 2.3.28-ubuntu22-arm</li>
</ul>
<h3>Change</h3>
<ul>
<li><strong>Dataproc on Compute Engine</strong>: Upgraded Apache Zookeeper to version <code>3.9.5</code> in image version <code>2.3</code>.</li>
</ul>
<h3>Fixed</h3>
<ul>
<li>Upgraded Dataproc Metastore Proxy to v0.0.79 to fix CVEs.</li>
<li>Fixed CVEs CVE-2026-24308 and CVE-2026-24281.</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 24, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_24_2026</id>
    <updated>2026-03-24T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_24_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.1.112-debian11, 2.1.112-rocky8, 2.1.112-ubuntu20, 2.1.112-ubuntu20-arm</li>
<li>2.2.80-debian12, 2.2.80-rocky9, 2.2.80-ubuntu22, 2.2.80-ubuntu22-arm</li>
<li>2.3.27-debian12, 2.3.27-ml-ubuntu22, 2.3.27-rocky9, 2.3.27-ubuntu22, 2.3.27-ubuntu22-arm</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 20, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_20_2026</id>
    <updated>2026-03-20T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_20_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.76</li>
<li>2.2.76</li>
<li>2.3.29</li>
<li>3.0.12</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 18, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_18_2026</id>
    <updated>2026-03-18T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_18_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.1.111-debian11, 2.1.111-rocky8, 2.1.111-ubuntu20, 2.1.111-ubuntu20-arm</li>
<li>2.2.79-debian12, 2.2.79-rocky9, 2.2.79-ubuntu22, 2.2.79-ubuntu22-arm</li>
<li>2.3.26-debian12, 2.3.26-ml-ubuntu22, 2.3.26-rocky9, 2.3.26-ubuntu22, 2.3.26-ubuntu22-arm</li>
</ul>
<h3>Fixed</h3>
<ul>
<li>Fixed CVEs CVE-2025-58057, CVE-2025-53864, CVE-2025-68161, CVE-2025-48924, and CVE-2025-33042.</li>
<li>Upgraded Dataproc Metastore Proxy to v0.0.78 to fix CVEs.</li>
<li>Default JDK is set to Temurin JDK - 11.0.30 in all <code>2.1</code>, <code>2.2</code> and <code>2.3</code> images.</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 13, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_13_2026</id>
    <updated>2026-03-13T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_13_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.75</li>
<li>2.2.75</li>
<li>2.3.28</li>
<li>3.0.11</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 11, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_11_2026</id>
    <updated>2026-03-11T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_11_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p><strong>Dataproc on Compute Engine:</strong> The following subminor image versions announced on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_08_2026">March 08, 2026</a> have been rolled back:</p>
<ul>
<li>2.1.110-debian11, 2.1.110-rocky8, 2.1.110-ubuntu20, 2.1.110-ubuntu20-arm</li>
<li>2.2.78-debian12, 2.2.78-rocky9, 2.2.78-ubuntu22, 2.2.78-ubuntu22-arm</li>
<li>2.3.25-debian12, 2.3.25-ml-ubuntu22, 2.3.25-rocky9, 2.3.25-ubuntu22, 2.3.25-ubuntu22-arm</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 08, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_08_2026</id>
    <updated>2026-03-08T00:00:00-07:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_08_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.1.110-debian11, 2.1.110-rocky8, 2.1.110-ubuntu20, 2.1.110-ubuntu20-arm</li>
<li>2.2.78-debian12, 2.2.78-rocky9, 2.2.78-ubuntu22, 2.2.78-ubuntu22-arm</li>
<li>2.3.25-debian12, 2.3.25-ml-ubuntu22, 2.3.25-rocky9, 2.3.25-ubuntu22, 2.3.25-ubuntu22-arm</li>
</ul>
<p><strong>Rollback Notice:</strong> These image versions were rolled back on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_11_2026">March 11, 2026</a>.</p>
<h3>Fixed</h3>
<ul>
<li>Fixed CVEs CVE-2025-58057, CVE-2025-53864, CVE-2025-68161, CVE-2025-48924 (partial), and CVE-2025-33042.</li>
<li>Upgraded Dataproc Metastore Proxy to v0.0.78 to fix CVEs.</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 06, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_06_2026</id>
    <updated>2026-03-06T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_06_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.74</li>
<li>2.2.74</li>
<li>2.3.27</li>
<li>3.0.10</li>
</ul>
<h3>Announcement</h3>
<p>Serverless for Apache Spark <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-3.0">3.0</a> runtime version:</p>
<ul>
<li>Updated Spark version to <a href="https://spark.apache.org/releases/spark-release-4-0-1.html">4.0.1</a>.</li>
<li>Updated BigQuery Spark Connector version to <a href="https://github.com/GoogleCloudDataproc/spark-bigquery-connector/releases/tag/0.44.0">0.44.0</a>.</li>
<li>Applied patch for <a href="https://issues.apache.org/jira/browse/SPARK-52601">SPARK-52601</a>.</li>
<li>Applied patch for <a href="https://issues.apache.org/jira/browse/SPARK-52614">SPARK-52614</a>.</li>
</ul>
<h3>Announcement</h3>
<p>Serverless for Apache Spark <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-1.2">1.2</a>, <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-2.2">2.2</a> and <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-2.3">2.3</a> runtime versions:</p>
<ul>
<li>Updated Delta Lake version to <a href="https://github.com/delta-io/delta/releases/tag/v3.2.1">3.2.1</a>.</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>March 03, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#March_03_2026</id>
    <updated>2026-03-03T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#March_03_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.73</li>
<li>2.2.73</li>
<li>2.3.26</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>February 27, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_27_2026</id>
    <updated>2026-02-27T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_27_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.72</li>
<li>2.2.72</li>
<li>2.3.25</li>
<li>3.0.9</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>February 24, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_24_2026</id>
    <updated>2026-02-24T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_24_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.160-debian10, 2.0.160-rocky8, 2.0.160-ubuntu18</li>
<li>2.1.109-debian11, 2.1.109-rocky8, 2.1.109-ubuntu20, 2.1.109-ubuntu20-arm</li>
<li>2.2.77-debian12, 2.2.77-rocky9, 2.2.77-ubuntu22, 2.2.77-ubuntu22-arm</li>
<li>2.3.24-debian12, 2.3.24-ml-ubuntu22, 2.3.24-rocky9, 2.3.24-ubuntu22, 2.3.24-ubuntu22-arm</li>
</ul>
<h3>Fixed</h3>
<p>Fixed an issue that could cause <code>gsutil</code> failures on clusters using custom images built from new base images. This problem occurred when the gcloud <code>core/universe_domain</code> property was not set in the build environment, resulting in a malformed <code>/etc/boto.cfg</code>. The image build process now ensures a correct default configuration, restoring proper Google Cloud Storage connectivity.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>February 20, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_20_2026</id>
    <updated>2026-02-20T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_20_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.71</li>
<li>2.2.71</li>
<li>2.3.24</li>
<li>3.0.8</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>February 19, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_19_2026</id>
    <updated>2026-02-19T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_19_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>Hyperdisks can now be added to a Dataproc cluster.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>February 15, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_15_2026</id>
    <updated>2026-02-15T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_15_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.159-debian10, 2.0.159-rocky8, 2.0.159-ubuntu18</li>
<li>2.1.108-debian11, 2.1.108-rocky8, 2.1.108-ubuntu20, 2.1.108-ubuntu20-arm</li>
<li>2.2.76-debian12, 2.2.76-rocky9, 2.2.76-ubuntu22, 2.2.76-ubuntu22-arm</li>
<li>2.3.23-debian12, 2.3.23-ml-ubuntu22, 2.3.23-rocky9, 2.3.23-ubuntu22, 2.3.23-ubuntu22-arm</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>February 13, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_13_2026</id>
    <updated>2026-02-13T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_13_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.70</li>
<li>2.2.70</li>
<li>2.3.23</li>
<li>3.0.7</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>February 06, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_06_2026</id>
    <updated>2026-02-06T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_06_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.69</li>
<li>2.2.69</li>
<li>2.3.22</li>
<li>3.0.5</li>
</ul>
<h3>Feature</h3>
<p><strong>Serverless for Apache Spark:</strong> Added support for removing <code>conscrypt</code> 
from Serverless for Apache Spark <code>2.3</code>+ runtimes using the
<a href="https://docs.cloud.google.com/dataproc-serverless/docs/concepts/properties#other_properties"><code>dataproc.artifacts.remove</code></a>
property .</p>
<h3>Feature</h3>
<p><strong>Dataproc on Compute Engine:</strong> <a href="https://docs.cloud.google.com/dataproc/docs/support/diagnose-clusters#snapshot_diagnostic_data_collection">Sharing snapshot diagnostic data</a>: Setting the <code>--tarball-access=GOOGLE_DATAPROC_DIAGNOSE</code> flag with the <a href="https://docs.cloud.google.com/sdk/gcloud/reference/dataproc/clusters/diagnose"><code>gcloud dataproc clusters diagnose</code></a> command shares all of the output Cloud Storage bucket contents with Google Cloud support if <a href="https://docs.cloud.google.com/storage/docs/uniform-bucket-level-access">uniform bucket-level access</a> is enabled on the output Cloud Storage bucket. If object-level access control is enabled on the output Cloud Storage bucket, only the generated diagnostic tar file is shared.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>February 05, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_05_2026</id>
    <updated>2026-02-05T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_05_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.158-debian10, 2.0.158-ubuntu18, 2.0.158-rocky8</li>
<li>2.1.107-debian11, 2.1.107-ubuntu20, 2.1.107-ubuntu20-arm, 2.1.107-rocky8</li>
<li>2.2.75-debian12, 2.2.75-ubuntu22, 2.2.75-ubuntu22-arm, 2.2.75-rocky9</li>
<li>2.3.22-debian12, 2.3.22-ml-ubuntu22, 2.3.22-rocky9, 2.3.22-ubuntu22, 2.3.22-ubuntu22-arm</li>
</ul>
<h3>Feature</h3>
<p>Zookeeper version upgraded to 3.9.4 in Dataproc image 2.3.22.</p>
<h3>Feature</h3>
<p>Parquet CLI version upgraded to 1.15.2 in 2.1 and 2.2 images.</p>
<h3>Feature</h3>
<p>Delta subminor version upgraded to 3.2.1 in Dataproc on Compute Engine image 2.2 and 2.3.</p>
<h3>Feature</h3>
<p>Apache Pig is now available in ARM images.</p>
<h3>Feature</h3>
<p>Added a new <code>dataproc:pypi.repository</code> property to customize the PyPI repository used for pip. The value can be a URL, or <code>google</code> to use a Google-hosted cache of PyPI, accessible without public internet connectivity. Starting in image version 3.1, <code>google</code> will be the default; to opt out and return to public PyPI, use the value <code>pypi</code>.</p>
<h3>Change</h3>
<p>Removed use of deprecated Hadoop configuration properties <code>fs.default.name</code> and <code>yarn.resourcemanager.system-metrics-publisher.enabled</code>.</p>
<h3>Fixed</h3>
<p>Fixed a bug in the ARM image that prevented connecting to a Dataproc Metastore instance with a gRPC protocol endpoint.</p>
<h3>Fixed</h3>
<p>Fixed the <code>spark.driver.extraClassPath</code> delimiter for the Jupyter SparkMonitor Listener.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>February 04, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#February_04_2026</id>
    <updated>2026-02-04T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#February_04_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p><strong>Upcoming Spark data lineage changes</strong> See the upcoming May,
2026 Dataproc and Serverless for Apache Spark release notes for an announcement
of a change that will automatically enable
<a href="https://docs.cloud.google.com/dataproc/docs/guides/spark-lineage">Dataproc Spark data lineage</a> and
<a href="https://docs.cloud.google.com/dataproc-serverless/docs/guides/data-lineage">Serverless for Apache Spark data lineage</a>
when you enable the Data Lineage API (see
<a href="https://docs.cloud.google.com/dataplex/docs/use-lineage#control-ingestion">Control lineage ingestion for a service</a>)
without requiring additional project, cluster, batch workload, or
interactive session settings.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>January 30, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_30_2026</id>
    <updated>2026-01-30T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_30_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>3.0.4</li>
</ul>
<h3>Announcement</h3>
<p><strong>Dataproc on Compute Engine:</strong> The following subminor image versions announced on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_24_2026">January 24, 2026</a> have been rolled back:</p>
<ul>
<li>2.0.157-debian10, 2.0.157-ubuntu18, 2.0.157-rocky8</li>
<li>2.1.106-debian11, 2.1.106-ubuntu20, 2.1.106-ubuntu20-arm, 2.1.106-rocky8</li>
<li>2.2.74-debian12, 2.2.74-ubuntu22, 2.2.74-ubuntu22-arm, 2.2.74-rocky9</li>
<li>2.3.21-debian12, 2.3.21-ml-ubuntu22, 2.3.21-rocky9, 2.3.21-ubuntu22, 2.3.21-ubuntu22-arm</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>January 24, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_24_2026</id>
    <updated>2026-01-24T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_24_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.157-debian10, 2.0.157-ubuntu18, 2.0.157-rocky8</li>
<li>2.1.106-debian11, 2.1.106-ubuntu20, 2.1.106-ubuntu20-arm, 2.1.106-rocky8</li>
<li>2.2.74-debian12, 2.2.74-ubuntu22, 2.2.74-ubuntu22-arm, 2.2.74-rocky9</li>
<li>2.3.21-debian12, 2.3.21-ml-ubuntu22, 2.3.21-rocky9, 2.3.21-ubuntu22, 2.3.21-ubuntu22-arm</li>
</ul>
<p><strong>Rollback Notice:</strong> These image versions were rolled back on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_30_2026">January 30, 2026</a>.</p>
<h3>Feature</h3>
<p>Upgraded the Delta subminor version to <code>3.2.1</code> in images <code>2.2</code> and <code>2.3</code>.</p>
<h3>Feature</h3>
<p>Apache Pig is now available in ARM images.</p>
<h3>Feature</h3>
<p>Added a new <code>dataproc:pypi.repository</code> property to customize the PyPI repository that <code>pip</code> uses.
The value can be a internet-accessible URL, or you can specify <code>google</code> to use a Google-hosted cache of PyPI, which
is accessible without public internet connectivity. Starting with Managed Service for Apache Spark image version
<code>3.1</code>, <code>google</code> will be the default (you can specify the <code>pypi</code> property value to use public PyPI instead of the default <code>google</code> 3.1 value).</p>
<h3>Change</h3>
<p>Removed the use of deprecated Hadoop configuration properties <code>fs.default.name</code> and
<code>yarn.resourcemanager.system-metrics-publisher.enabled</code>.</p>
<h3>Fixed</h3>
<p>Fixed a bug in the ARM image that prevented connecting to a Dataproc Metastore instance with a
gRPC protocol endpoint.</p>
<h3>Fixed</h3>
<p>Fixed the <code>spark.driver.extraClassPath</code> delimiter for the Jupyter SparkMonitor Listener.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>January 23, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_23_2026</id>
    <updated>2026-01-23T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_23_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.67</li>
<li>2.2.67</li>
<li>2.3.20</li>
<li>3.0.3</li>
</ul>
<h3>Fixed</h3>
<ul>
<li>Applied patch for <a href="https://issues.apache.org/jira/browse/SPARK-48292">SPARK-48292</a> in Serverless for Apache Spark <a href="https://docs.cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-1.2">1.2</a> and <a href="https://docs.cloud.google.com/dataproc-serverless/docs/concepts/versions/spark-runtime-2.2">2.2</a> runtimes.</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>January 13, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_13_2026</id>
    <updated>2026-01-13T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_13_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p><strong>Dataproc on Compute Engine:</strong> The following subminor image versions announced on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_06_2026">January 06, 2026</a> have been rolled back:</p>
<ul>
<li>2.0.156-debian10, 2.0.156-ubuntu18, 2.0.156-rocky8</li>
<li>2.1.105-debian11, 2.1.105-ubuntu20, 2.1.105-ubuntu20-arm, 2.1.105-rocky8</li>
<li>2.2.73-debian12, 2.2.73-ubuntu22, 2.2.73-ubuntu22-arm, 2.2.73-rocky9</li>
<li>2.3.20-debian12, 2.3.20-ml-ubuntu22, 2.3.20-rocky9, 2.3.20-ubuntu22, 2.3.20-ubuntu22-arm</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>January 09, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_09_2026</id>
    <updated>2026-01-09T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_09_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.66</li>
<li>2.2.66</li>
<li>2.3.19</li>
<li>3.0.1</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>January 06, 2026</title>
    <id>tag:google.com,2016:dataproc-release-notes#January_06_2026</id>
    <updated>2026-01-06T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_06_2026"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.156-debian10, 2.0.156-ubuntu18, 2.0.156-rocky8</li>
<li>2.1.105-debian11, 2.1.105-ubuntu20, 2.1.105-ubuntu20-arm, 2.1.105-rocky8</li>
<li>2.2.73-debian12, 2.2.73-ubuntu22, 2.2.73-ubuntu22-arm, 2.2.73-rocky9</li>
<li>2.3.20-debian12, 2.3.20-ml-ubuntu22, 2.3.20-rocky9, 2.3.20-ubuntu22, 2.3.20-ubuntu22-arm</li>
</ul>
<p><strong>Rollback Notice:</strong> These image versions were rolled back on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#January_13_2026">January 13, 2026</a>.</p>
<h3>Feature</h3>
<p>Added a new property <code>dataproc:pypi.repository</code> to customize the PyPI repository used for pip. The value can be a URL, or <code>google</code> to use a Google-hosted cache of PyPI, accessible without public internet connectivity. Starting in image version 3.1, <code>google</code> will be the default; to opt out and return to public PyPI, use the value <code>pypi</code>.</p>
<h3>Change</h3>
<p>Removed use of deprecated Hadoop configuration properties <code>fs.default.name</code> and <code>yarn.resourcemanager.system-metrics-publisher.enabled</code>.</p>
<h3>Fixed</h3>
<p>Fixed the <code>spark.driver.extraClassPath</code> delimiter for the Jupyter SparkMonitor Listener.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>December 22, 2025</title>
    <id>tag:google.com,2016:dataproc-release-notes#December_22_2025</id>
    <updated>2025-12-22T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#December_22_2025"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://cloud.google.com/dataproc-serverless/docs/concepts/versions/dataproc-serverless-versions">Serverless for Apache Spark runtime versions</a>:</p>
<ul>
<li>1.2.65</li>
<li>2.2.65</li>
<li>2.3.18</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>December 20, 2025</title>
    <id>tag:google.com,2016:dataproc-release-notes#December_20_2025</id>
    <updated>2025-12-20T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#December_20_2025"/>
    <content type="html"><![CDATA[<h3>Breaking</h3>
<p>The following <strong>Dataproc on Compute Engine</strong> subminor image versions,
released on <a href="https://docs.cloud.google.com/dataproc/docs/release-notes#December_05_2025">December 05, 2025</a>, are now <strong>blocklisted</strong>:</p>
<ul>
<li>2.2.72-debian12, 2.2.72-ubuntu22, 2.2.72-ubuntu22-arm, 2.2.72-rocky9</li>
<li>2.3.19-debian12, 2.3.19-ubuntu22, 2.3.19-ubuntu22-arm, 2.3.19-ml-ubuntu22, 2.3.19-rocky9</li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>December 05, 2025</title>
    <id>tag:google.com,2016:dataproc-release-notes#December_05_2025</id>
    <updated>2025-12-05T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#December_05_2025"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters#supported-dataproc-image-versions">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.155-debian10, 2.0.155-ubuntu18, 2.0.155-rocky8</li>
<li>2.1.104-debian11, 2.1.104-ubuntu20, 2.1.104-ubuntu20-arm, 2.1.104-rocky8</li>
<li>2.2.72-debian12, 2.2.72-ubuntu22, 2.2.72-ubuntu22-arm, 2.2.72-rocky9</li>
<li>2.3.19-debian12, 2.3.19-ubuntu22, 2.3.19-ubuntu22-arm, 2.3.19-ml-ubuntu22, 2.3.19-rocky9</li>
</ul>
<p><strong><aside><strong></strong>Blocklist update:</aside></strong> The <code>2.2.72</code> and <code>2.3.19</code>
subminor image versions were blocklisted on
<a><href="https: dataproc="" docs="" docs.cloud.google.com="" release-notes#december_20_2025"="">December 20, 2025</href="https:></a>.</p>
<h3>Feature</h3>
<p><code>dataproc-ml</code> is available by default in <code>2.3</code> <code>-ml</code> images.</p>
]]>
    </content>
  </entry>

  <entry>
    <title>December 04, 2025</title>
    <id>tag:google.com,2016:dataproc-release-notes#December_04_2025</id>
    <updated>2025-12-04T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#December_04_2025"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p><strong>Serverless for Apache Spark:</strong> Runtime version <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-release-3.0">3.0</a> is now generally available. This version simplifies onboarding, improves reliability, reduces startup latency, and adds support for Spark 4.0.</p>
<ul>
<li><p>Features and improvements:</p>
<ul>
<li>Regional and multi-zonal workloads are used by default to increase obtainability of compute resources</li>
<li>Faster startup than previous runtimes</li>
<li>Fast resource cleanup that allows faster release of VPC IPs after workload completion</li>
<li>End-user credentials are used for all workloads by default</li>
<li>New <code>bigquery</code> Spark catalog, pre-configured for out-of-the-box BigQuery
native table interactions</li>
<li>New Spark Serverless-specific IAM roles</li>
<li>New <code>dataproc-rm.googleapis.com</code> API enablement is required</li>
</ul></li>
</ul>
]]>
    </content>
  </entry>

  <entry>
    <title>November 22, 2025</title>
    <id>tag:google.com,2016:dataproc-release-notes#November_22_2025</id>
    <updated>2025-11-22T00:00:00-08:00</updated>
    <link rel="alternate" href="https://docs.cloud.google.com/dataproc/docs/release-notes#November_22_2025"/>
    <content type="html"><![CDATA[<h3>Announcement</h3>
<p>New <a href="https://docs.cloud.google.com/dataproc/docs/concepts/versioning/dataproc-version-clusters">Dataproc on Compute Engine subminor image versions</a>:</p>
<ul>
<li>2.0.154-debian10, 2.0.154-ubuntu18, 2.0.154-rocky8</li>
<li>2.1.103-debian11, 2.1.103-ubuntu20, 2.1.103-ubuntu20-arm, 2.1.103-rocky8</li>
<li>2.2.71-debian12, 2.2.71-ubuntu22, 2.2.71-ubuntu22-arm, 2.2.71-rocky9</li>
<li>2.3.18-debian12, 2.3.18-ubuntu22, 2.3.18-ubuntu22-arm, 2.3.18-ml-ubuntu22, 2.3.18-rocky9</li>
</ul>
<h3>Change</h3>
<p><strong>Dataproc on Compute Engine</strong>: Updated Cloud Storage connector:</p>
<ul>
<li>Updated to <code>3.1.10</code> in image version <code>2.3.18</code>.</li>
<li>Updated to <code>3.0.15</code> in image version <code>2.2.71</code>.</li>
<li>Updated to <code>hadoop3-2.2.30</code> in image versions <code>2.1.103</code> and <code>2.0.154</code>.</li>
</ul>
]]>
    </content>
  </entry>

</feed>
