DocumentationCommerceTools

ACP2E-4908: Large catalogs cause excessive Redis or Valkey memory usage

Last update: November 2, 2026

CREATED FOR:

  • Experienced
  • Admin
  • Developer

The ACP2E-4908 patch fixes the issue where large catalogs could cause excessive memory use in Valkey (or Redis) because separate layout cache entries were created for each product in each store view. This patch is available when the Quality Patches Tool (QPT) 1.1.83 is installed. The patch ID is ACP2E-4908. Please note that this issue is scheduled to be fixed in Adobe Commerce 2.4.10.

Affected products and versions

The patch is created for Adobe Commerce version:

  • Adobe Commerce (all deployment methods) 2.4.8-p4

Compatible with Adobe Commerce versions:

  • Adobe Commerce (all deployment methods) >=2.4.8 <2.4.10
NOTE
The patch might become applicable to other versions with new Quality Patches Tool releases. To check if the patch is compatible with your Adobe Commerce version, update the magento/quality-patches package to the latest version and check the compatibility on the Quality Patches Tool: Search for patches page. Use the patch ID as a search keyword to locate the patch.

Issue

Large catalogs can cause excessive memory use in Redis or Valkey because separate layout cache entries are created for each product in each store view.

Steps to reproduce:

  1. Configure Valkey (or Redis) as the default cache backend.

  2. Edit app/etc/env.php and set the cache.frontend.default block to:

    code language-none 
    'default' => [
    'id_prefix' => '061_',
    'backend_options' => [
        'server' => 'valkey',          // your Valkey/Redis hostname
        'port' => '6379',
        'database' => '1',
        'password' => '',
        'compress_data' => '0',        // key recipe — leaves XML uncompressed
        'persistent' => 0,
        'read_timeout' => 10,
        'connect_retries' => 2,
    ],
    'backend' => 'redis',
],

    Then flush: bin/magento cache:flush.

  3. Generate the medium performance fixture:

    code language-none 
    bin/magento setup:performance:generate-fixtures \
    setup/performance-toolkit/profiles/ce/medium.xml

    This produces 3 websites × 3 store groups × 3 store views, 24K simple + 640 configurable products, 33 attribute sets, 50K orders. Expect 10 min on a fast machine.

  4. Reset Valkey db1 and warm up:

    code language-none 
    valkey-cli -h <host> -n 1 FLUSHDB
bin/magento cache:flush

  5. Crawl product pages across stores. Build a sample list and hit each URL:

    code language-none 
    # Build 200 random product URLs across stores 1–4
mysql -e "SELECT CONCAT(s.code,'|',u.request_path)
          FROM url_rewrite u JOIN store s ON u.store_id=s.store_id
          WHERE u.entity_type='product'
            AND u.request_path NOT LIKE 'catalog/product/%'
            AND s.store_id BETWEEN 1 AND 4
          ORDER BY RAND() LIMIT 200;" magento > urls.txt

while IFS='|' read -r STORE URL; do
    curl -s -o /dev/null "http://<base-url>/${URL}?___store=${STORE}"
done < urls.txt

  6. Observe the LAYOUT cache footprint:

    code language-none 
    valkey-cli -h <host> -n 1 --scan --pattern "*LAYOUT_FRONTEND_STORE*" | wc -l
valkey-cli -h <host> INFO memory | grep used_memory_human

    Scale up the crawl (or use large.xml/extra_large.xml and a real load tool like siege/wrk) to reach the customer’s 10K-keys / 1.7 GB regime.

Expected results:

Performance is not impacted when compression is enabled.

Actual results:

  • The LAYOUT_FRONTEND_STORE Redis keys consume a significant amount of space.
  • On a fresh medium-fixture install, after 200 product hits: approximately 564 LAYOUT_FRONTEND_STORE keys, and used_memory_human of approximately 20 MB (vs. approximately 10 MB baseline before the crawl) — approximately 2.8 LAYOUT cache keys per product page hit, growing linearly with unique products visited × stores.

Apply the patch

To apply individual patches, use the following links depending on your deployment method:

To learn more about Quality Patches Tool, refer to:

recommendation-more-help
commerce-operations-help-tools