Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    Squid3 cachemgr cache_mem info different than config file(BUG?)

    Scheduled Pinned Locked Moved pfSense Packages
    2 Posts 2 Posters 764 Views
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • perikoP
      periko
      last edited by

      Hi.

      I'm working in my new server running pfsense 2.1.5 x64, the goal is to setup squid+bind.

      Now, tunning my server which have 8GBRAM, this will serve +/- 100 hungry users.

      I setup cachemgr to see details of my squid setup, but for some reasons the value:

      cache_mem

      I different from squid config file generate by pfsense system.

      I want to let squid for example 5120GB RAM I want him to be happy and serve faster hot objects.

      Example, if I write the value = 5120 show me:

      squid -k parse
      2014/11/07 10:58:47| Processing: cache_mem 5120 MB

      Now, cachemgr show me:

      cache_mem 1073741824 bytes

      Different value.

      I will trust squid console command, some else with this error?

      Thanks.

      Necesitan Soporte de Pfsense en México?/Need Pfsense Support in Mexico?
      www.bajaopensolutions.com
      https://www.facebook.com/BajaOpenSolutions
      Quieres aprender PfSense, visita mi canal de youtube:
      https://www.youtube.com/c/PedroMorenoBOS

      1 Reply Last reply Reply Quote 0
      • T
        Tikimotel
        last edited by

        It is not a good idea to offer more than 50% of your 8192 MB RAM for squid cache.

        Squid uses RAM for in transit objects for the disk cache.
        This process needs around 100MB of RAM per GB of disk cache.
        It is dynamically allocated depending on the object stored in the cache, so this is not an exact science.

        If you have a large disk cache set, 100GB for instance.
        You will need, when the disk cache is in full use, 10GB of RAM!

        Depending on your disk cache, set the memory cache to such a size you know will remain available after the disk cache is in full use.
        (other process also use RAM, bind etc.)

        I use around 25% of my RAM for the memory cache and use the 100MB per GB rule to size the disk cache.

        1 Reply Last reply Reply Quote 0
        • First post
          Last post
        Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.