Failing Jobs finds the background jobs that keep failing and explains why, in plain language: what the stuck item is, how big it is, which worker runs it, and the line of code where it dies. Report only: no queue item is released or deleted.
When cron "runs" but nothing gets done, the cause is usually one queue item that always fails. Cron always takes the oldest item first, so that one item blocks everything behind it, often for weeks, while the logs only say "out of memory". Failing Jobs puts the item, the worker and the crash on one card.
What it finds
- Stuck queue jobs: queues whose oldest item is older than a set number of hours (24 by default), or that have an item claimed right now. The card names the item (when it is an entity: type, bundle, title and the size of its large text fields), how many items wait and since when, how many point at deleted content, and the worker (class, file, cron time).
- PHP fatal errors (out of memory, time limit, uncaught errors) from the PHP and web server error logs, grouped by file and line, with the line of code, how often and when, and the request (web cron keys are masked). A crash in the same module as a stuck job's worker is shown on that job's card as the reason it never finishes.
- No memory limit for PHP on the command line, so a manual
drush cronorqueue:runcan take all of the server's memory.
A real example, as the card reads it: Content item 12497 'Document Outline': body 14.0 MB, 6.9 million characters, about 41,200 sentences. 8 items in the queue; stuck since 2026-09-28 (9.9 days ago). Worker: Extract metadata for migration nodes. Its runs die: out of memory (1 GB limit) at NonAiMetadataExtractor.php:148, during web cron.
Severity Finding High Stuck queue whose worker's module crashes Medium Stuck queue; fatal error; no command-line memory limit Low Stuck queue whose items all point at deleted contentBuilt for government and regulated sites
Relevant NIST SP 800-53 Rev. 5 controls: SI-4 System Monitoring, SI-11 Error Handling, SC-5 Denial-of-service Protection, SC-6 Resource Availability. The report reads queues and logs only; it changes nothing, and web cron keys are never shown.
Use
- Reports » Failing jobs (
/admin/reports/failing-jobs). Needs the View failing jobs permission. - Drush:
drush failing-jobs, which exits non-zero while a stuck job keeps crashing;drush failing-jobs --fields=title,details,code,fix --format=yamlfor the details. - Error logs are read from the end (the last 2 MB of each): PHP's own
error_log, nginx, Apache, PHP-FPM and Acquia's log folders by default, or your own list infailing_jobs.settings:error_logs. Logs the web server cannot read are listed as such.
Getting started
composer require drupal/failing_jobs:1.0.x-dev@devand enable the module.- Give View failing jobs to the people who look after the site.
- Open Reports » Failing jobs, or run
drush failing-jobs.
Requirements
Drupal ^10.3 || ^11 || ^12. No other modules. Drush 12.5 or later for the command (optional).
Related
Drupal Cleanup includes the same failing-jobs report alongside its other cleanup reports. Install Failing Jobs on its own when that report is all you need.
Ask your assistant about Failing Jobs
Check compatibility with your Drupal and PHP version, maintenance and security coverage, from current release data. How it works
Depends on
Dependencies of the latest stable release
No dependencies recorded for this project.
Required by
Tracked projects that depend on this one
No tracked projects depend on this one yet.
Activity
Releases
| Version | Type | Core | PHP | Release date | |
|---|---|---|---|---|---|
| 1.0.x-dev | Dev | 10–11 | Nov 21, 2026 |