If your production site is returning "Temporarily Unavailable" pages, or has been reported to be down when using using New Relic uptime monitoring within your Acquia subscription, or the site is extremely slow, there are various actions you can take.

• Common causes
  Common causes of "Temporarily Unavailable" pages on production environments include but are not limited to:
  • Transitory resource limits such as insufficient CPU or DB resources, etc. due to high traffic (either expected, or from unusual robot crawling or DDoS attacks) and/or application complexity.
  • Code and/or data changes or bugs that trigger PHP exceptions or errors.
  • Resolution will vary depending on the situation, and may include rolling back changes to a site (restoring older configuration, code and/or a database backup), adjusting PHP memory limits, blocking traffic via a Web Application Firewall (WAF) through Acquia Edge or other methods, etc.

To request help for severely-impacted production sites, create a Support ticket with critical urgency for immediate assistance.

• Check the Acquia status page

  The Acquia status page reports issues that Acquia is aware of that affect our customers’ applications, including Amazon Web Services issues, external DNS issues, or other incidents. In some cases, your applications may be down due to global issues that are beyond your control. Incidents that cause applications to be down are treated as the highest level of severity and Acquia works continuously until they are resolved. Acquia updates the status page regularly until the incident is resolved.

• Check the Cloud Platform interface Stack Metrics and Production environment Overview pages

  The Cloud Platform interface provides valuable information about your application’s health.

  The Stack Metrics page reports on your application’s CPU, memory, and storage usage. If your environments are running on Acquia’s Cloud Classic infrastructure and your disk storage space is full, your applications might become impaired.

  The Overview page for an environment shows a task log. Examine the task log and look for failed tasks. Check whether there have been any changes to the code or database in your Production environment that might have caused your application to go down.

• Verify your logs and errors
  
  Logs and log streaming should be used to determine the root cause. Access logs will show the real HTTP status code at the backend, and matching logs from other sources (PHP Errors, Drupal Watchdog, etc) can be found using the X-Request-Id value which will match all logged items for the same request.
  
  We also recommend using the New Relic Application Monitoring tool included with your subscription to verify

• Address out of memory errors

  Use the following steps to resolve these errors:

  • Investigate memory consumption to identify major memory-consuming processes.
  • Reduce the frequency of cron jobs and remove the jobs that are unnecessary.
  • Adjust the PHP memory limit per environment to ensure your application can load.
  • Avoid overriding environment-wide memory limits as this can destabilize your infrastructure.

  For detailed information on these steps, visit Memory leaks.

• Evaluate resource allocation and caching

  If your application experiences high traffic or service interruptions due to insufficient caching, evaluate and adjust the available resources. Consider increasing memory or optimizing cache settings through the Acquia Cloud UI. For load-related issues, it might be necessary to check cron jobs and explore performance improvements using caching solutions. Implement these changes during off-peak hours to minimize disruptions.

• Check the logs

  Check your infrastructure error and access logs for errors. For information about what to look for, see Searching the error logs to troubleshoot problems.

  You can download your logs from the Logs page. For more information, see About Cloud Platform logging.

• Contact Acquia Support

  If your Acquia subscription entitles you to file support tickets, you can create a Support ticket. In general, if your application is down due to a global issue noted on the Acquia Status page, there is no need to create a ticket. Contact Acquia Support if an issue is particular to your application.

  For issues involving a production site that is down or severely impacted, file a case with a Critical urgency. This ensures immediate attention from our emergency response support team. Specify the nature of your production issue clearly when submitting the ticket.

  Acquia responds to critical urgency requests promptly, regardless of the time of day.

What to do if your application is down