Pimcore cache issues and recommendations


Hi guys,

This is our current setup running Pimcore 5: 3 EC2 instances (m5.xlarge) x 3 docker containers each. We have a shared folder between them (everything inside var/), all assets (images, css, js files) are stored on cloudfront. Sometimes after deploy, bin/console becomes unresponsive and server goes into 500 error. The only solution we found so far is to manually delete the cache folder (rm -rf var/cache/) and then both website and bin/console come back again, the cache being rebuilt.

Except this, we are using New Relic to monitor our website and the results are not great. App server is fine, but page view is not (please check attached screenshot). I was wondering if any of you can help with recommendations - what are we doing wrong? I imagine Pimcore should be capable of handling much more traffic.



Hello Alexdan,

For bin/console issue, please check the permissions before running the bin/console commands.
Also, let us know what medium/service are you using for shared storage.

For performance, below are some considerations.

  • Push your cache configuration to Redis. This will also help in rectifying the first issue you mentioned.
  • Try increasing memory_limit parameter in php.ini.
  • You can also check for database optimization.



Hi @RajatToshniwal,

could you be more specific about what you mean by

Push your cache configuration to Redis.

Am I right to assume you mean reconfiguring the cache currently living in <pimcore_root>/var/cache/ to be stored in Redis instead?

If so, I’d be really curious how to do this, as I haven’t been able to figure out how this could be done.


Hi Alexdan,

Please use this link “https://pimcore.com/docs/5.x/Development_Documentation/Development_Tools_and_Details/Cache/index.html”.
It will push all the cache (except symfony) to redis.

Another option could be to put the cache on the local disk (EBS) instead of shared storage. The only drawback of this approach is you can not use the clear-cache option provided on the admin panel. For clearing the cache, you will have to run “bin/console” command on all the servers.


Hi @RajatToshniwal,

We are already caching everything we can into Redis according to Pimcore’s documentation, but after we increased our traffic from 3 to 30 page views per minute, the performance got worse. I’m afraid that after we will migrate our old website (500 page views per minute), we will be in big trouble. Anything else we can do?

We cannot use full page cache unfortunately because we have different prices per country and other dynamic data.



I think you need to provide a few more information:

  • What are you deploying and how?
  • What technology is the shared folder? Is there fast read/write access available?

So far it looks to me as a general Symfony topic (since container rebuild seems to be the problem?), not so much a Pimcore specific thing. Maybe you find there some more information?