drip-cli

drip-cli

A command-line tool for managing Drip job queues. Supports PostgreSQL, MariaDB, and SQLite. Includes both a plain-text output mode, JSON output for scripting, and an interactive TUI browser built with charm.clj.

Requires Java 21+.

drip-cli — Drip job queue management CLI

Usage:
  drip-cli [global-opts] <command> [command-opts]

Global options:
  --url <jdbc-url>    JDBC URL (or set DRIP_JDBC_URL env var)
  --format <fmt>      Output format: table (default) | json

Commands:
  jobs list           List jobs
  jobs get            Get a single job by ID (--id <id>)
  jobs cancel         Cancel a job (--id <id>)
  jobs retry          Retry a job (--id <id>)
  jobs discard        Discard a job (--id <id>)
  jobs delete         Delete a single job (--id <id>)
  jobs delete-many    Delete multiple jobs by filter (--states <s1,s2>)
  jobs tui            Interactive TUI browser (--refresh <seconds>, default 5)

  queues list         List queues
  queues pause        Pause a queue (--name <n>)
  queues resume       Resume a queue (--name <n>)

  migrate             Run database migrations

jobs list options:
  --state <s>              Single state filter
  --states <s1,s2,...>     Multiple states (OR)
  --kind <k>               Single kind filter
  --kinds <k1,k2,...>      Multiple kinds (OR)
  --queue <q>              Single queue filter
  --queues <q1,q2,...>     Multiple queues (OR)
  --limit <n>              Max results (default 100)
  --after <id>             Cursor: return jobs with id < after
  --created-after <iso>    ISO-8601 instant lower bound on created_at
  --created-before <iso>   ISO-8601 instant upper bound on created_at
  --scheduled-after <iso>  ISO-8601 instant lower bound on scheduled_at
  --scheduled-before <iso> ISO-8601 instant upper bound on scheduled_at

jobs delete-many options:
  --states <s1,s2,...>     States to delete (required)
  --kinds <k1,k2,...>
  --queues <q1,q2,...>
  --created-before <iso>
  --finalized-before <iso>

Examples:
  drip-cli --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs list --state failed
  drip-cli --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs tui
  drip-cli --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs retry 42
  drip-cli --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs retry --id 42
  DRIP_JDBC_URL='jdbc:...' drip-cli jobs list --format json | jq .
  drip-cli --url 'jdbc:...' migrate

Installation

Build an uberjar:

clojure -T:build uber
# produces target/drip-cli.jar

Run:

java -jar target/drip-cli.jar --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs list

Or via alias during development:

clojure -M:run --url 'jdbc:postgresql://localhost/mydb?user=pg&password=pg' jobs list

Connection

Provide the JDBC URL via --url or the DRIP_JDBC_URL environment variable. The DB type (postgres/mariadb/sqlite) is detected automatically from the URL.

export DRIP_JDBC_URL='jdbc:postgresql://localhost/mydb?user=app&password=secret'
drip-cli jobs list

Usage

drip-cli [global-opts] <command> [command-opts]

Global options:
  --url <jdbc-url>    JDBC URL (or set DRIP_JDBC_URL env var)
  --format <fmt>      Output format: table (default) | json

Commands:
  jobs list           List jobs
  jobs get            Get a single job by ID (--id <id>)
  jobs cancel         Cancel a job (--id <id>)
  jobs retry          Retry a job (--id <id>)
  jobs discard        Discard a job (--id <id>)
  jobs delete         Delete a single job (--id <id>)
  jobs delete-many    Delete multiple jobs by filter (--states <s1,s2>)
  jobs tui            Interactive TUI browser (--refresh <seconds>, default 5)

  queues list         List queues
  queues pause        Pause a queue (--name <n>)
  queues resume       Resume a queue (--name <n>)

  migrate             Run database migrations

Commands

Jobs

List jobs:

drip-cli jobs list
drip-cli jobs list --state failed
drip-cli jobs list --states running,retryable
drip-cli jobs list --kind send_email --limit 50
drip-cli jobs list --queue priority --format json | jq '.[] | .id'
drip-cli jobs list --created-after 2026-01-01T00:00:00Z

Filter options:

Option	Description
--state <s>	Single state: available, running, completed, retryable, discarded, cancelled, scheduled
--states <s1,s2>	Multiple states (OR)
--kind <k>	Single kind filter
--kinds <k1,k2>	Multiple kinds (OR)
--queue <q>	Single queue filter
--queues <q1,q2>	Multiple queues (OR)
--limit <n>	Max results (default 100)
--after <id>	Cursor pagination: return jobs with id < after
--created-after <iso>	ISO-8601 lower bound on created_at
--created-before <iso>	ISO-8601 upper bound on created_at
--scheduled-after <iso>	ISO-8601 lower bound on scheduled_at
--scheduled-before <iso>	ISO-8601 upper bound on scheduled_at

Get a job:

drip-cli jobs get --id 42
drip-cli jobs get --id 42 --format json

Transition a job:

drip-cli jobs cancel  --id 42
drip-cli jobs retry   --id 42
drip-cli jobs discard --id 42
drip-cli jobs delete  --id 42

Bulk delete:

# Delete all completed and cancelled jobs older than a date
drip-cli jobs delete-many --states completed,cancelled --finalized-before 2026-01-01T00:00:00Z

# Delete discarded jobs of a specific kind
drip-cli jobs delete-many --states discarded --kinds send_email

Options for delete-many:

Option	Description
--states <s1,s2>	States to delete (required)
--kinds <k1,k2>	Restrict to these kinds
--queues <q1,q2>	Restrict to these queues
--created-before <iso>	Only delete created before this instant
--finalized-before <iso>	Only delete finalized before this instant

Interactive TUI:

drip-cli jobs tui
drip-cli jobs tui --refresh 10   # auto-refresh every 10 seconds (default: 5)

The TUI shows a scrollable jobs table with auto-refresh. Job detail view shows all fields grouped by section (identity, state, timestamps, args, metadata, errors) with stack trace previews for failed jobs.

Keyboard shortcuts:

Key	Action
j / ↓ / ctrl+n	Move down
k / ↑ / ctrl+p	Move up
ctrl+d / pgdn	Page down
ctrl+u / pgup	Page up
g / home	Jump to top
G / end	Jump to bottom
enter	View job detail
r	Refresh now (resets auto-refresh timer)
a	Toggle auto-refresh on/off
q	Quit / back
esc	Back to list (from detail)

Queues

drip-cli queues list
drip-cli queues list --format json
drip-cli queues pause  --name default
drip-cli queues resume --name default

Migrations

Run pending migrations (idempotent):

drip-cli migrate

JSON output

All list and get commands accept --format json. Output goes to stdout and is suitable for piping to jq:

# IDs of all running jobs
drip-cli jobs list --state running --format json | jq '.[].id'

# Count jobs by state
drip-cli jobs list --limit 1000 --format json | jq 'group_by(.state) | map({state: .[0].state, count: length})'

# Find jobs that errored more than 3 times
drip-cli jobs list --state discarded --format json | jq '.[] | select(.attempt > 3)'

Building

# Uberjar
clojure -T:build uber

# Run from source
clojure -M:run --help

License

Copyright © 2026 Max Penet Distributed under the Mozilla Public License Version 2.0