Drupal performance check

--- name: performance-check description: Analyze Drupal site performance and caching configuration arguments: "[module_path]" --- # Drupal Performance Check Analyze site performance configuration and identify optimization opportunities. Usage: /performance-check [module_path] If a module path is provided, focus analysis on that module's code. Otherwise, check site-wide performance settings and scan all custom modules. ## Performance Analysis Steps ### 1. Check Caching Configuration ``bash # Page cache drush config:get system.performance cache.page.max_age # CSS/JS aggregation drush config:get system.performance css.preprocess drush config:get system.performance js.preprocess # Twig cache (should NOT be disabled in production) drush php:eval "var_dump(\$settings['cache']['bins']['render'] ?? 'default');" ` **Production should have:** - cache.page.max_age > 0 (e.g., 3600 for 1 hour) - css.preprocess = true - js.preprocess = true - No cache bins disabled ### 2. Check Cache Backends `bash # What cache backends are configured? drush php:eval "print_r(\$settings['cache']['default'] ?? 'database');" # List all cache bins drush cache:list ` **Recommendations:** - Use Redis or Memcache for production (not database) - Consider separate backends for different cache bins ### 3. Review BigPipe Status `bash drush pm:list --filter=big_pipe drush config:get big_pipe.settings ` BigPipe should be enabled for authenticated user performance. ### 4. Check Database Performance `bash # Show slow query log (if enabled) drush watchdog:show --filter="type=php" --severity=warning --count=20 # Check database size drush sql:query "SELECT table_name, round(((data_length + index_length) / 1024 / 1024), 2) AS 'Size (MB)' FROM information_schema.tables WHERE table_schema = DATABASE() ORDER BY (data_length + index_length) DESC LIMIT 10;" ` Look for: - Large cache tables (may need pruning) - Large watchdog table (configure log rotation) - Orphaned data ### 5. Analyze Cache Metadata in Custom Code Search for missing cache metadata: `bash # Find render arrays without cache metadata grep -r "#markup" modules/custom/ | grep -v "#cache" # Find entity queries that might need caching grep -r "getQuery" modules/custom/ grep -r "entityQuery" modules/custom/ ` Every render array should include #cache with appropriate: - tags - What data does this depend on? - contexts - What varies this output? (user, url, permissions) - max-age - How long is this valid? ### 6. Check for N+1 Query Patterns `bash # Find potential N+1 patterns (loading entities in loops) grep -r "->load(" modules/custom/ | head -20 grep -r "entity_load" modules/custom/ | head -20 # Find queries inside foreach grep -rB5 "->load(" modules/custom/ | grep -A5 "foreach" ` **Fix N+1 patterns:** - Use loadMultiple() instead of load() in loops - Use entity reference fields with proper loading - Pre-load entities before iteration ### 7. Views Performance `bash # List all views drush views:list # Check views with lots of results drush views:analyze ` Common views issues: - Missing caching on views - No pagination on large result sets - Inefficient filters ### 8. Check for Static Caching `bash # Services without caching expensive operations grep -r "public function get" modules/custom/src/Service/ | head -20 ` Expensive operations (API calls, complex queries) should use static caching: `php protected array $cache = []; public function getExpensiveData(string $id): array { if (!isset($this->cache[$id])) { $this->cache[$id] = $this->computeExpensiveData($id); } return $this->cache[$id]; } ` ### 9. Check Cron and Queue Performance `bash # Last cron run drush core:cron --verbose 2>&1 | head -5 # Queue status drush queue:list ` For heavy processing: - Use Queue API instead of cron - Process queues with separate workers - Consider ultimate_cron for better cron management ### 10. Frontend Performance `bash # Check library definitions find modules/custom -name "*.libraries.yml" -exec cat {} \; # Look for inline JS/CSS (should use libraries) grep -r "<script" modules/custom/templates/ grep -r "<style" modules/custom/templates/ ` Best practices: - All JS/CSS via libraries.yml (enables aggregation) - Defer non-critical JS - Use critical CSS for above-the-fold ## Performance Report Format ### Critical Issues - Page cache disabled - CSS/JS aggregation off in production - Database cache backend in production - N+1 query patterns ### Warnings - Missing cache metadata on render arrays - Views without caching - Large database tables - Inline JS/CSS ### Optimization Opportunities - Consider Redis/Memcache - Implement lazy builders for personalized content - Add static caching to expensive services - Move heavy processing to queues ## Quick Wins Checklist ``bash # Enable all standard performance settings drush config:set system.performance cache.page.max_age 3600 drush config:set system.