Quarkus Community Call - 2026-07-21

[cescoffier]

Hello,

Yesterday, we had our Quarkus community call, the last one before the August break. We restart on September 1 with a session on Bazel, followed by Gradle on September 15.

Here is a short summary. You can find the complete minutes (including the recording) on https://docs.google.com/document/d/1TgFZsuOQo9qZ4CnQII5LHhQVgMC6YsVMs1UJIAJooyM/edit?tab=t.0#heading=h.vsr1oshl0610. The recording is available at https://www.youtube.com/watch?v=RaQlgXXHVr8.

Recap:

James Cobb proposed organizing a QuarkusCon, a streamed online conference targeting November 2026 to avoid conflicts with Devoxx and end-of-year holidays. The format is still open (curated talks, call for papers, or a mix), and the team will gauge community interest and speaker availability via Zulip and GitHub Discussions. If you are interested in attending or presenting, keep an eye on both channels.

On the release front, Quarkus 3.40 is confirmed as the final 3.x release. Patch releases (3.40.1, 3.40.2, etc.) will continue, but there will be no 3.41 or beyond, the next LTS is Quarkus 4.0. The team is also actively triaging a high volume of security reports while maintaining SLAs, with the next periodic release run planned for August 11.

Quantum-safe TLS support has landed in both 3.40 and 4.0 with two modes: a JVM-based mode using Java 27 early access (cross-platform, zero-config) and an OpenSSL 3.5 mode (Linux only, faster and more memory-efficient). Neither mode works in native image yet, that is planned for Quarkus 4. A blog post explaining configuration will follow.

Holly Cummins presented a deep dive on migrating quarkus.io from Jekyll to Roq. The migration covers the main site only (not /extensions or /benchmarks). Holly built a comprehensive Playwright and Lighthouse test suite — the site's first automated testing, and developed reusable conversion scripts (~10,000 lines) rather than relying on one-shot AI conversion, allowing iterative improvement while picking up upstream changes for free. The PR touches ~42,000 files. Key results: CI test execution is 42% faster with Roq (2.8 min vs. 4 min with Jekyll), Lighthouse performance scores are consistently better thanks to Roq's asset bundling, and the effort surfaced 2 Qute defects, ~6 Roq defects, and ~5 feature requests — validating the "eat our own dog food" approach. The main challenge was the semantic gap between Liquid and Qute around variable scoping and mutability. The migration PR is ready to merge with a short ~30-minute code freeze, and a hackit script is available for reconciling the ~30 open PRs against quarkus.io. Guillaume Smet raised a question about whether Qute re-processes AsciiDoc variable content as templates, which will be clarified with Martin Kouba.