Skip to content

DOC-7315 Add Resque, Asynq, and Hangfire integration pages - #4603

Merged
andy-stark-redis merged 1 commit into
mainfrom
DOC-7315-add-resque-asynq-and-hangfire-integration-docs
Nov 17, 2026
Merged

andy-stark-redis merged 1 commit into
mainfrom
DOC-7315-add-resque-asynq-and-hangfire-integration-docs

Conversation

@andy-stark-redis

@andy-stark-redis andy-stark-redis commented Nov 17, 2026 •

Copy link
Copy Markdown
Contributor

Summary

  • Adds three integration pages under content/integrate/: Resque (Ruby), Asynq (Go), and Hangfire (.NET), following the Celery page's shape.
  • Links them from the job-queue Ecosystem section, removes the Spring Batch bullet (its "Redis job repository" claim is wrong), and rewords the Hangfire bullet to say its Redis storage is a community package.
  • Follow-up to DOC-7313 and DOC-7314; part of epic DOC-7308; closes DOC-7315.

Verification

Every snippet was run against Redis 7.2.7 with Resque 3.2.0, Asynq 0.26.0 (Go 1.27), and Hangfire.Core 1.8.25 with Hangfire.Redis.StackExchange 1.12.0 (.NET 9). The kill-recovery behavior differs for each:

  • Resque: never re-runs the job. A later worker marks it failed (Resque::DirtyExit) once the dead worker's heartbeat is older than the prune interval (300 s by default).
  • Asynq: lease-based. The task stays active for about a minute, then retries; a new server re-ran it about 73 s after the kill.
  • Hangfire: the job waits in a dequeued list until the storage InvisibilityTimeout passes (30 min by default). With 10 s, a new server re-ran it 62 s after the kill.

For review

  • Hangfire. The core package has no Redis storage. The vendor's Redis package (Hangfire.Pro.Redis) is part of the paid Hangfire Pro, so the page documents the community Hangfire.Redis.StackExchange (LGPL v3, third-party, no official support) and says which is which. Pricing is not stated.
  • Spring Batch. spring-batch-core 6.0.6 and 6.1.0-RC1 contain only a MongoDB job repository, so I removed the bullet, as agreed.
  • Bull is untouched. It is in maintenance mode and points to BullMQ; that's a separate wording decision.
  • Not run (and not claimed): Resque retry and scheduling plugins, Asynq unique and periodic tasks and Redis Cluster, the Hangfire dashboard and recurring jobs. Upstream links are site roots.
  • Weights (17 to 19) follow RQ and Dramatiq (15, 16) in DOC-7314 Add RQ and Dramatiq integration pages #4602.

🤖 Generated with Claude Code


Note

Low Risk
Documentation-only changes with no runtime or application code impact.

Overview
Adds three new integration guides—Resque (Ruby), Asynq (Go), and Hangfire (.NET)—aligned with existing pages like Celery: setup, enqueue/work patterns, Redis key layout, and behavior when a worker or server is killed.

The job-queue Ecosystem section now links to these docs instead of external URLs, drops the Spring Batch Redis job-repository entry, and clarifies that Hangfire’s Redis backing is via a community storage package (not core Hangfire).

Reviewed by Cursor Bugbot for commit f23866a. Bugbot is set up for automated code reviews on this repo. Configure here.

Adds content/integrate/resque, asynq, and hangfire, and links them from the job-queue
use case's Ecosystem section. Also removes the Spring Batch bullet: it claimed a "Redis
job repository", but spring-batch-core 6.0.6 and 6.1.0-RC1 contain only a MongoDB one.
The Hangfire bullet now says its Redis storage is a community package. Bull is left
alone; it is in maintenance mode and points to BullMQ.

Every snippet was run against Redis 7.2.7 with Resque 3.2.0 (Ruby 4.0.6), Asynq 0.26.0
(Go 1.27.2), and Hangfire.Core 1.8.25 with Hangfire.Redis.StackExchange 1.12.0 (.NET 9).
The three behave differently when a worker is killed with kill -9. Resque never re-runs
the job: with the default 60 second heartbeat and 300 second prune interval a later
worker marks it failed as Resque::DirtyExit, and with 2 and 6 seconds that happened 5
seconds after the kill. Asynq holds a lease on each active task, so the task stayed
active for about a minute, moved through retry and pending, and a new server re-ran it
about 73 seconds after the kill. Hangfire leaves the job in a dequeued list until the
storage InvisibilityTimeout (30 minutes by default) passes; with 10 seconds, a new server
re-ran it 62 seconds after the kill. Asynq and Hangfire both poll for scheduled and
retried jobs, about every 5 and 15 seconds, so short delays are not accurate.

Hangfire core has no Redis storage. The vendor's Redis package, Hangfire.Pro.Redis, is
part of the paid Hangfire Pro, and the page documents the community package
Hangfire.Redis.StackExchange (LGPL v3, listed on Hangfire's extensions page, no official
support) and says which is which. Pricing is deliberately not stated.

Learned: Resque, Asynq, and Hangfire recover a killed worker's job in three different ways (never, lease expiry, invisibility timeout); Hangfire's Redis storage is a community package
Constraint: Kill-recovery and polling numbers come from single-server runs on the versions named; recheck if a library changes how it recovers or polls
Directive: Do not say Hangfire supports Redis without naming the community package, and do not state Hangfire Pro pricing
Gaps: Resque plugins for retry and scheduling not run; Asynq unique tasks, periodic tasks, and Redis Cluster not run; Hangfire dashboard, recurring jobs, and Hangfire.Pro.Redis not run; links to upstream docs are site roots, not deep pages
Recheck: when Resque 4, Asynq 1.0, or Hangfire 2.0 ships
Ticket: DOC-7315
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

github-actions Bot commented Nov 17, 2026 •

Copy link
Copy Markdown
Contributor

DOC-7315

@github-actions

Copy link
Copy Markdown
Contributor

🧠 Redis Memory

Found 5 related items from repository history (5 new this commit):

Memory updated at f23866a

@dwdougherty dwdougherty left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@andy-stark-redis andy-stark-redis self-assigned this Nov 17, 2026
@andy-stark-redis andy-stark-redis added the ecosystem Docs for external Redis integrations and resources label Nov 17, 2026
@andy-stark-redis
andy-stark-redis merged commit 32a1728 into main Nov 17, 2026
114 checks passed
@andy-stark-redis
andy-stark-redis deleted the DOC-7315-add-resque-asynq-and-hangfire-integration-docs branch November 17, 2026 16:15
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ecosystem Docs for external Redis integrations and resources

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants