Troubleshooting Magento

Diagnose Magento 2 / Adobe Commerce on TurboStack - find logs, fix maintenance mode, cache, indexers, Elasticsearch, cron and Varnish 503 issues.

Most Magento problems have a few common causes: stale caches, invalid indexers, a stopped cron, a search-engine connection, or the site stuck in maintenance mode. This page shows where to look and how to fix the common cases. Many fixes use Magento's CLI (bin/magento) or the diagnostic tool n98-magerun2, both available from your app root.

Where to find the logs

Your Magento app lives at /var/www/<user>/public_html (the vhost serves its pub/ subdirectory). The most useful logs are:

Component Where
Magento application /var/www/<user>/public_html/var/log/ (system.log, exception.log, debug.log)
Magento cron var/log/magento.cron.log and var/log/setup.cron.log
Reports / fatal errors /var/www/<user>/public_html/var/report/
Nginx access/error the host's web server logs (under the system user's home / /var/log/)
PHP-FPM the PHP-FPM pool log for the app's runtime
Database (MySQL) the MySQL error/slow-query log
Varnish the Varnish service log

All files under var/log/ are rotated weekly. You can also use the host's Health tab for resource graphs and check recent deploys under History.

Common issues

Symptom Likely cause Fix
Site shows "service unavailable" / maintenance page Maintenance mode left enabled (often after a failed deploy) bin/magento maintenance:status, then bin/magento maintenance:disable
Old content, changed config or prices not showing Stale cache bin/magento cache:flush (and cache:clean)
Catalog/prices wrong, "indexer invalid" warning Indexers out of date bin/magento indexer:status, then bin/magento indexer:reindex
Storefront errors, empty search results Elasticsearch/OpenSearch down or misconfigured Verify the service is running on Services; re-check the search engine host and index prefix in admin
Emails, reindex, or scheduled jobs never run Cron not running Confirm the cron jobs exist for the system user and read var/log/magento.cron.log; run bin/magento cron:run manually to test
503 errors from the storefront Varnish backend (PHP/nginx) down, or wrong Varnish hosts Check PHP-FPM/nginx are up; verify Magento's http-cache-hosts and Varnish backend port (8080) settings
White screen / blank pages, permission errors Wrong file permissions or stale generated code Fix ownership of var/, generated/, pub/static; clear with rm -rf generated/* var/cache/* then setup:di:compile
Admin/storefront very slow Running in developer mode in production bin/magento deploy:mode:set production
Indexer stuck or failing repeatedly Corrupted indexer state bin/magento indexer:reset, then indexer:reindex; in production set indexers to cron-driven with indexer:set-mode schedule
Missing or stale CSS/JS in the storefront Static content not deployed bin/magento setup:static-content:deploy -f (-f forces the deploy even in developer mode)
Stock quantities wrong after orders or returns Inventory reservations inconsistent bin/magento inventory:reservation:list-inconsistencies, then inventory:reservation:create-compensations
Locked out of the admin Too many failed admin logins bin/magento admin:user:unlock <username>

A troubleshooting workflow

  1. Check Health - rule out resource exhaustion (CPU, memory, disk) and confirm the host is up.
  2. Read the relevant log - start with var/log/exception.log and var/log/system.log, plus the nginx/PHP-FPM logs for HTTP-level errors.
  3. Check the last deploy in History - if a recent change broke the site, revert or re-publish a known-good revision via Publishing.
  4. Verify services are running - on Services confirm MySQL, Elasticsearch/OpenSearch, Redis, and Varnish are all up; then re-run cache:flush and indexer:reindex and re-test.

Getting help

If you are still stuck, reach out via Support, or work through the platform-wide Troubleshooting guide for issues that are not Magento-specific.

© 2026 Hosted Power — TurboStack