The Solr PMC is pleased to announce the release of Apache Solr 9.11.0.
Solr is the blazing-fast, open source, multi-modal search platform built on Apache Lucene. It powers full-text, vector, analytics, and geospatial search at many of the world's largest organizations. Other major features include Kubernetes and docker integration, streaming, highlighting, faceting, and spellchecking. Solr 9.11.0 is available for immediate download at: <https://solr.apache.org/downloads.html> ### Solr 9.11.0 Release Highlights: * Added new range field types — IntRangeField, LongRangeField, FloatRangeField, and DoubleRangeField — for storing and querying numeric ranges, plus an experimental numericRange query parser. * Extended nested/block document support: the parent and child query parsers gained a parentPath local parameter (childPath also added) that automatically derives the correct filter from document nesting, so queries no longer need to reference the internal nest path field directly. Also added a root-document query shortcut for nested document queries. Index sorting now works correctly with nested documents. * Added hybrid search based on reciprocal rank fusion (RRF) via a new combined query search handler, merging multiple query strategies into one ranked result set. Also supports result collapsing and query elevation. * Added support for multiple dense vectors per document (nested vectors), enabling multi-valued vector search. * Added MergeOnFlushMergePolicy, merging small segments together at commit, reducing segments to search. * Added ContentHashVersionProcessor, an update processor avoiding index churn when re-adding documents with identical content. * Expanded streaming expression rollup with new metrics: missing-value counts, count-distinct, percentile, standard deviation. * Added support for stored fields in Export Writer output, not just docValues. * Added an UPGRADECOREINDEX core admin action to upgrade an index format in place without reindexing. * The Luke handler now aggregates results across all shards by default. * Improved multiThreaded=true query performance when a docset is needed, such as faceting. * JWT authentication's blockUnknown setting now defaults to true, blocking unauthenticated requests by default; it previously defaulted to false despite documentation stating otherwise. Deployments relying on pass-through authentication must now explicitly set blockUnknown to false. * Added an allowZkHosts setting to restrict which ZooKeeper connection strings a cross-collection join or streaming expression may target. The local SolrCloud ensemble is always allowed. * Removed LocalTikaExtractionBackend from the extraction module (SolrCell); extracting content via a remote Tika Server is now the only option. * And more features/improvements. * Fixed MANY bugs. Please refer to the Upgrade Notes in the Solr Ref Guide for information on upgrading from previous Solr versions: < https://solr.apache.org/guide/solr/9_11/upgrade-notes/solr-upgrade-notes.html > Please read CHANGELOG.md for a full list of new features, changes and bugfixes: <https://solr.apache.org/9_11_0/changes/Changes.html> ### Thanks to all contributors: Abhishek Umarjikar, Alessandro Benedetti, Andrzej Bialecki, Andy Webb, Arup Chauhan, Bartosz Fidrysiak, David Smiley, Dmitrii Tikhonov, Eric Pugh, Francois Huaulme, Gaurav Tuli, Houston Putman, Idan Tepper, Jalaz Kumar, James Vanneman, Jan Høydahl, Jason Gerlowski, Li Enlai, Luke Kot-Zaniewski, Mark Miller, Matthew Biscocho, Olivier Boudet, Pierre Salagnac, Puneet Ahuja, Puneet Sharma, Rahul Goswami, Rajat Raghav, Ravi Ranjan Jha, Renato Haeberli, Samuel Verstraete, Sanjay Kumar Yadav, Serhiy Bzhezytskyy, Shiming Li, Shrey Narayan, Shubham Ranjan, Sonu Sharma, Tony Panza, Vishnu Priya Chandra Sekar, William Wallace, Xinyao Zhang, ZhenyuLi, hossman, khushjain, r4mercur
