Failed Jobs

Holo-JS supports a shared failed-job store for async drivers. By default, config/queue.ts uses a database-backed failed-job table named failed_jobs.

List failed jobs

npx holo queue:failed

Use this to inspect failed jobs and copy the failed-job record ID for retry or deletion.

Retry failed jobs

Retry one failed job:

npx holo queue:retry 01JZ123456789ABCDEFGHJKMNP

Retry every stored failed job:

npx holo queue:retry all

Delete one failed job record

npx holo queue:forget 01JZ123456789ABCDEFGHJKMNP

This removes the failed-job record without retrying it.

Flush the failed-job store

npx holo queue:flush

Use this only when you intentionally want to clear every stored failed job record.

Failed-job storage config

The default scaffold keeps failed jobs in the database:

import { defineQueueConfig } from '@holo-js/queue'
export default defineQueueConfig({
  failed: {
    driver: 'database',
    connection: 'default',
    table: 'failed_jobs',
  },
})

Disable failed-job persistence when you do not want to keep failed async jobs:

import { defineQueueConfig } from '@holo-js/queue'
export default defineQueueConfig({
  failed: false,
})

When failed jobs are stored

• sync jobs run inline and do not go through the async failed-job store.
• redis workers can persist failed jobs.
• database workers can persist failed jobs.

If you want failed-job commands to work, keep the failed-job store enabled and generate the failed jobs table.

Continue

• Database Tables
• Workers