Support bound Redis job dispatch with a semaphore - #12
Open
jbaxendale-ut wants to merge 1 commit into
Open
Conversation
A long-running application may need a Redis worker to share a bounded capacity with other work without changing the existing task-parent behavior. Previously the server had no semaphore-aware dispatcher lifecycle or permit ownership model. Accept an Async::Semaphore parent, hold a permit from blocking fetch through job completion, and keep only one blocking fetch in flight. The default task-parent behavior remains unchanged. A single dispatcher launcher preserves the required Async lifecycle in each mode, and a single processing cleanup path releases permits exactly once. On a failed fetch or heartbeat, stop dispatching rather than spin, retain active workers for cancellation, and release remaining permits on shutdown. This gives any Redis-backed AsyncJob host a predictable concurrency boundary without coupling the processor to application-specific scheduling. Signed-off-by: Jeff Baxendale <jbaxendale@usertesting.com>
jbaxendale-ut
force-pushed
the
fix/semaphore-dispatcher-default
branch
from
September 3, 2026 20:22
8c1729d to
8bc63a6
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add support for using an
Async::Semaphoreto bound Redis job processing, which we needed to support rate limiting an external API (ultimately reserving interview capacity for multiple users)When a semaphore is supplied, a permit is held from the blocking fetch through job completion and only one blocking fetch is in flight. The server also retains active workers for cancellation, releases permits during shutdown, and stops dispatching if dequeue or heartbeat recovery has failed rather than entering a reconnect/logging loop.
Disclaimer: developed with the help of Codex as I was investigating latency issues in an LLM-moderated interview flow in our application. Let me know if this seems in appropriate, but wanted to offer my changes (running in production now) upstream see also socketry/async-job-adapter-active_job#29
Hosts that need a bounded worker can apply a shared concurrency limit without changing queue semantics or allowing jobs to accumulate outside the limit.
Types of Changes
Contribution