Ai

Onyx v3.2.10 Adds Admin Controls, Metrics, and Routing Fixes

Onyx v3.2.10 Adds Admin Controls, Metrics, and Routing Fixes

Onyx v3.2.10 delivers a focused update aimed at operational control, observability, and deployment reliability. The release adds new admin settings for file upload size and token limits, expands Helm support with Prometheus metrics endpoints for Celery workers, and includes several routing and frontend fixes that improve platform stability.

What Changed

The headline feature in v3.2.10 is new administrator control over file upload size and token limits. This gives teams more flexibility to tune the platform for different workloads, governance requirements, and infrastructure constraints without relying on code-level changes.

On the infrastructure side, the Helm deployment now supports Prometheus metrics ports and Services for Celery workers. That is an important addition for teams running Onyx in production Kubernetes environments, as it improves monitoring coverage and makes worker-level visibility easier to integrate into existing observability stacks.

There are also multiple nginx routing updates in this release. One fix routes /scim/* requests to the API server, while another adds a routing rule for non-API backend paths. Together, these changes should reduce deployment edge cases and help backend traffic reach the correct services more consistently.

In the indexing layer, the Vespa and OpenSearch index function was refactored to use Iterable, which points to internal cleanup and maintainability improvements. The release also includes a performance-related change to permission sync start time handling and a new ods trace devtools update, alongside an ods upgrade from 0.7.1 to 0.7.2.

Frontend users also get a smaller but useful fix: popover content now avoids overflowing on small screens, improving usability on constrained display sizes.

Finally, the version includes dependency maintenance updates such as bumps for Handlebars, Picomatch, and Brace Expansion, helping keep the web and backend-adjacent components current.

Why It Matters

This version stands out less for a single major feature and more for tightening operational readiness. Admin-configurable limits make the platform easier to govern, while Prometheus metrics for Celery workers strengthen production monitoring and capacity planning.

The nginx routing fixes are especially important for teams running identity workflows or more complex backend traffic patterns, because misrouting issues at that layer can create hard-to-diagnose failures. By addressing those paths directly, v3.2.10 improves deployment reliability.

Combined with indexing refactors, performance tweaks, frontend polish, and dependency maintenance, this release looks like a practical platform-hardening update for organizations running Onyx in active production environments.

Official Source: https://github.com/onyx-dot-app/onyx/releases/tag/v3.2.10