Problem with memcache and W3 Total Cache

  • Resolved Sobolle

    (@sobolle)


    4 years, 1 month ago

    Has anyone faced a memory overflow issue with memcache? (or a similar problem, we don’t know for sure)
    We use virtual hosting with memcache installed. We are using the W3 Total Cache caching plugin, version 2.1.1
    At night, google was indexing 15,000 pages of the site. Then something happened to the cache, and the resource load on the virtual hosting became above 600% of the maximum allowable level, and this continued until we noticed this in the morning and restarted memcache.
    How to deal with this?

    Through this script, you can view the memcache status: https://dsmotors.ru/memcache-details-en.php

    The page I need help with: [log in to see the link]

Viewing 3 replies - 1 through 3 (of 3 total)
  • Plugin Contributor Marko Vasiljevic

    (@vmarko)

    Hello @sobolle

    I am sorry about the issue you experienced and I am happy to assist you with this.
    The latest version for CentOS 7 is Memcached-1.4.15-10.el7_3.1.x86_64, which is the same version as reported.
    The Memcached website has a newer version: https://memcached.org/files/memcached-1.6.9.tar.gz So you might want to try upgrading because version 1.4.15 is from (2012-9-3)
    You can try configuring a memory limit for the Memcached server.

    -m, --memory-limit=<num>
Use <num> MB memory max to use for object storage; the default is 64 megabytes.

    Have you experienced this kind of issue before?
    Make sure to keep an eye on the resource load and see if the issue continues to persits.
    Thanks!

    Thread Starter Sobolle

    (@sobolle)

    Thanks you.

    Previously, the virtual hosting support service described the situation as follows:
    The load is caused by the intensive work of the Memcached socket process:
    170372 - S 195% / usr / bin / memcached -v -u motors -s /home/motors/tmp/memcached.sock -a 600 -P /home/motors/tmp/memcached.pid -m 1024 -I 1024
    This behavior of memcached is also observed when a large number of connections are made to the socket. This may be due to incorrect operation of the W3 Total Cache caching plugin.
    Later, when I sent them your answer, they wrote:
    Changing the version of Memcached within the framework of shared hosting is not provided. This update will affect the projects of all clients of the site. Since there were no such problems previously, most likely the complexity is caused by the automatic updating of the W3 Total cache plugin or one of the other plugins, which led to their conflict. This assumption is confirmed by the release of a new version of the W3 Total cache plugin on February 22nd.

    Are there any other recommendations on this?

    • This reply was modified 4 years ago by Sobolle.
    Plugin Contributor Marko Vasiljevic

    (@vmarko)

    Hello @sobolle

    Thank you for the information.
    Yes, that may be the case especially with the limited resources in shared hosting.
    We have not touched anything when it comes to Memcached in the latest W3TC release so it may be just a temporary hiccup on the server.
    The recommendation is to upgrade your hosting plan to a dedicated server so you can avoid any limitations.
    Thanks!

Viewing 3 replies - 1 through 3 (of 3 total)
  • The topic ‘Problem with memcache and W3 Total Cache’ is closed to new replies.

Tags