Knowledge Base - Articles

Posted 13 Apr 2018


I noticed that all webform submissions are missing from our Test database after a staging deployment. Is this expected behavior during a deploy?



Yes, this is expected behavior. Webform submissions are included in cache_form tables and our documentation on Scrubbing sensitive data from staged sites specifies that the following items are scrubbed:

  • Caching information (including CDNs, if used)


Posted 12 Apr 2018

When installing the Drupal 8 memcache module, the order of the steps is important, if the Memcache module is not installed before updating the settings.php file you could get errors like:  "non-existent cache service"  or “ServiceNotFoundException: error.”

Posted 11 Apr 2018

In Drupal 8, the caching system is much more flexible. Instead of caching the entire page (as in Drupal 7) , in Drupal 8 caches are saved in ‘bins’ which can be specifically set to use either the database or memcache, or other caching systems.  We recommend that you use memcache for most cache bins, this is preferable to local cache ( for example PHP APCu) for multi-tier stacks since it makes cached items available to all of the PHP/FPM layers 

Posted 09 Apr 2018

Known Issues

Acquia Support uses New Relic as a tool to help diagnose website issues. One part of the tool is a small Javascript agent that is injected into the page, that allows New Relic to give data about the end user's experience.

However, there maybe be issues caused by this javascript injection to a third party service used, or, a validation error, like this one from Google's AMP service:

The tag 'script' is disallowed except in specific forms.


Posted 09 Apr 2018


Some, or many, page loads are taking just longer than fifteen seconds to complete (as seen in the web server access.log file, or on Acquia Cloud, in the php_time value of the drupal-requests.log file). Memcached is in use.

Posted 29 Mar 2018

 * Pause between deleting large files, or use patience when making large numbers of deletes. 

Our file system is set to an active/active configuration which means that both the primary and replicated volume keep a table of reads and writes. Making a large number of changes swiftly, or making changes that push the I/O capacity of the instance can cause the file system replication to essentially lose track of the updates that are pending, and lead to unexpected results. Out of Memory issues and networking problems can also lead to impairment.

Posted 29 Mar 2018

PHP OPcache  Caches pre-compiled bytecode. This is quite a bit faaster than using PHP to load and parse each script for each and every HTTP request. This cache needs to be large enough to store compiled code for your applications PHP scripts, but not so large that it keeps you from using available RAM for other operations. 

Clear sign that this is too small:

Posted 29 Mar 2018

The PHP APC User Cache is the portion of PHP APC cache that is not included in Zend OPcache. It is rarely used in Drupal 7 and for most drupal 7 applicaitons (apc.shm_size= 8M is ok). We recommend that you increase this amount for Drupal 8.x applications either to (apc.shm_size=16 or 32M)

Clear sign that the APC User cache is set too small:

Posted 26 Mar 2018

When preparing to launch a website, or debugging problems with a site that's already live, sometimes it can be helpful to bypass CDN and proxy layers when requesting content from the site by sending those web requests directly to a specific IP address without using the site's public DNS records.

Posted 26 Mar 2018

During the process of customizing the storage backends used by Drupal's caching system (e.g.: when configuring Memcache on your Drupal site) it can be helpful to check which backends Drupal is actually using for each individual cache bin. The Drush commands below can be used for debugging purposes to quickly see which storage backends are being used for each cache bin in use by your Drupal 7 or 8 site:

Drupal 8:

Contact supportStill need assistance? Contact Acquia Support