Quarkus SmallRye Fault Tolerance @Bulkhead missing queue size configuration
25.6K reputation · 27 May 2025, 09:44 UTC
Quarkus SmallRye Fault Tolerance @Bulkhead missing queue size configuration
The goal is to determine how to bound the number of pending requests when a @Bulkhead concurrency limit is reached, without relying on an implicit unbounded queue that could exhaust memory.
Currently the @Bulkhead annotation does not expose a maxQueueSize attribute, and the underlying Vert.x worker pool does not provide a Quarkus‑specific property to limit the internal queue length. Adjusting the global Vert.x worker pool size (quarkus.vertx.worker-pool-size) affects all Vert.x‑based components, making isolation difficult.
Is there a way to set a maximum queue length for @Bulkhead without altering the shared Vert.x worker pool?
Does Quarkus plan to add a maxQueueSize attribute to the @Bulkhead annotation?
What recommended practices exist to bound pending tasks while keeping the bulkhead’s queue behavior isolated?
vertx