Alfresco drops Solr, Pristy stays open with Solr 9Solr 9

The search engine behind Pristy, the open-source document management suite published by Jeci, is now called Pristy Search Service and has just reached version 1.1.0. It is a free/open-source fork of Alfresco Search Services, ported to Apache Solr 9, which Jeci has been developing for several months — and which takes on new importance now that Alfresco Community has removed Apache Solr from its distribution.

Why now?

In July 2026, Alfresco Community 26.2 made a decision with heavy consequences: Apache Solr is no longer a supported search engine, on any edition (Alfresco Community 26.2 release notes). No more fixes, no more security updates, no more compatibility guarantee.

Its replacement, Alfresco Search Community, relies on Elasticsearch or OpenSearch, which are free and technically solid engines themselves. But the Alfresco connector that feeds them is shipped as binaries, with no published source code. Hyland says it intends to open it later; to date, neither the source code nor the final licence is available.

For any organisation, and even more so for a public body, this is no small matter. The search engine reads every single one of your documents. If it is closed, you cannot audit what it does, fix a vulnerability yourself, or prove how your platform works. You also have to run an Elasticsearch or OpenSearch cluster alongside it.

Pristy is built on Alfresco Community. We therefore had two options: follow this closed path, or keep search 100% free/open-source. We chose to stay open — and rather than stick with an ageing Solr, we brought it up to date.

The details of this choice and the before/after table are in our Pristy Core 26.1 vs. Alfresco Community 26.2 comparison.

What changes compared to Alfresco Search Services

The technical foundation is in place — unmodified Solr 9.10.1 and Lucene 9.12.3, Java 17, and an architecture that now separates indexing from search itself, so that each can be restarted and sized independently. Version 1.1.0 has been tested with Alfresco Community 26.2 and put through load tests on large volumes. It also brings:

• around 10% disk space saved, thanks to index compression;
• a new reranking that brings the most recent documents forward in results;
• a new real-time indexing monitoring interface;
• mTLS encryption of every link between Alfresco, Solr and the indexing trackers.

For the details of this architecture and what it concretely changes for running Pristy, head to our dedicated technical article.

This work reaches beyond Pristy: Hyland, the publisher of Alfresco, has presented it twice on its blog — a dedicated article in June 2026, then its reference guide to Alfresco search in September — and has added it as an option in its community Docker installer. The code is public on GitHub and GitLab.

A fork maintained for the whole community

This fork is not a one-off project. Jeci will keep maintaining it for the whole Alfresco community, not just for Pristy: tracking Apache Solr releases, fixes, and code that stays public. Any organisation that wants to keep a 100% free/open-source Alfresco search can rely on it.

And like any open-source project, it also moves forward thanks to the people who use it: feedback, bug reports and contributions are welcome — see the repository’s contribution guide.

Further reading

• Pristy Search Service, Pristy’s search engine on Solr 9 — architecture and concrete benefits for running Pristy
• Pristy Core 26.1 vs. Alfresco Community 26.2 — the full comparison, status and timeline included
• Pristy Core 26.0 vs. Alfresco Community 25.3 — the comparison for the version in production