Solid Queue Lite

A minimal, zero-build web interface for Solid Queue.

The official mission_control-jobs engine is great, but it brings along Turbo, Stimulus, and expects a standard Rails asset pipeline. If you run an API-only app, use a modern JS framework, or just want to avoid frontend dependencies in your infrastructure tooling, Solid Queue Lite provides the same operational visibility without the build-step baggage.

Add the engine to your host application's Gemfile:

gem "solid_queue_lite"

Install dependencies, then run the engine migration:

bundle install
bin/rails solid_queue_lite:install

The installer copies the engine migration into the host app and creates config/initializers/solid_queue_lite.rb if it does not already exist.

Run the migration separately, or let the installer do it for you:

bin/rails db:migrate
# or
bin/rails solid_queue_lite:install MIGRATE=1

Mount the engine behind your application's own authentication boundary:

# config/routes.rb
authenticate :user, ->(user) { user.admin? } do
	mount SolidQueueLite::Engine => "/ops/jobs"
end

The engine root renders the dashboard at /ops/jobs, and the jobs index is available at /ops/jobs/jobs.

• Ruby 3.1+
• Rails 7.1
• Solid Queue 1.x

gem "solid_queue_lite"

Use the configuration block to scope all dashboard reads in multi-tenant deployments:

# config/initializers/solid_queue_lite.rb
SolidQueueLite.configure do |config|
	config.tenant_scope = lambda do |relation|
		relation.where(queue_name: Current.account.solid_queue_prefix)
	end
end

The lambda receives the Active Record relation before it is queried.

The engine ships with SolidQueueLite::TelemetrySamplerJob, which writes aggregate queue snapshots into solid_queue_lite_stats and prunes samples older than 7 days.

If you use Solid Queue recurring tasks, schedule the sampler in your host application's recurring configuration:

# config/recurring.yml
production:
	solid_queue_lite_telemetry:
		class: SolidQueueLite::TelemetrySamplerJob
		schedule: every 5 minutes
		queue: default

Without a scheduler process, the dashboard still renders, but the historical charts stay empty until samples are written.

The reusable logic now lives under lib/solid_queue_lite, so you can call the same APIs from a Rails console:

SolidQueueLite::Telemetry.sample!
SolidQueueLite::Telemetry.backfill!
SolidQueueLite::Telemetry.dashboard_data(range_key: "24h")

SolidQueueLite::Jobs.list(state_key: "failed", page: 1, per_page: 50)
SolidQueueLite::Jobs.find(123)

SolidQueueLite::Processes.index_data
SolidQueueLite::Processes.pause_queue!("background")

To force an immediate current telemetry snapshot after upgrading, run:

bin/rake solid_queue_lite:telemetry:backfill

The historical scheduled_count and success_count series cannot be reconstructed exactly for old rows because Solid Queue does not retain that event history. The backfill task writes or refreshes the current snapshot immediately, so upgraded installs do not need to wait for the next scheduled sample.

• Zero Asset Pipeline: The UI is built with raw HTML, Pico.css, and Alpine.js loaded via CDN. It adds nothing to your app's frontend build.
• Database Safe: Standard job dashboards often kill primary databases with naive COUNT(*) queries on massive tables. This engine strictly uses database metadata (e.g., pg_class in Postgres) for approximate counting to prevent sequential scans and lock contention.
• Built-in Telemetry: Solid Queue doesn't store historical data. This engine includes a lightweight, self-pruning background job that snapshots queue sizes, latency, and error rates, giving you 24-hour trends without needing an external APM.
• Multi-tenant Support: Exposes a simple configuration block to scope the dashboard's database queries, allowing you to isolate job visibility per tenant.

• Monitor active Supervisors, Workers, and Dispatchers.
• Filter and inspect Ready, In-Progress, Scheduled, and Failed jobs.
• View job arguments (JSON), full stack traces, and execute individual or bulk retries/discards.
• Configurable auto-refresh that automatically pauses when you interact with the UI to prevent state loss.
• Monitor recurring task schedule, last run time, next run time, and latest status from a dedicated dashboard tab.