Thanks Joshua for the quick reply. Guess this means using mod_mem_cache will not give me the best cache hit/miss ratio as compared to mod_disk_cache ?
Is the way I have configured apache wrong ? Is it possible to run only 1 instance of httpd on a box to maximize the cache hit/miss ratio ? Would this approach cause other problems ? Is there any other MPM I should look at ?
I will look into mod_disk_cache as well..but can you please help me with the above questions...Thanks in advance...
Is the way I have configured apache wrong ? Is it possible to run only 1 instance of httpd on a box to maximize the cache hit/miss ratio ? Would this approach cause other problems ? Is there any other MPM I should look at ?
I will look into mod_disk_cache as well..but can you please help me with the above questions...Thanks in advance...
-----Original Message-----
From: [EMAIL PROTECTED]
To: users@httpd.apache.org
Sent: Fri, 1 Sep 2006 6:55 PM
Subject: Re: [EMAIL PROTECTED] mod_mem_cache question...
On 9/1/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>
>
> Hi,
>
> I have an Apache 2.2.2 setup on a Redhat box. Caching is implemented using
> mod_mem_cache.
> Apache was built to use worker MPM. Here are the relevant sections from
> httpd.conf
>
> #
> # mod_cache/mod_mem_cache settings
> #
> <IfModule cache_module>
> <IfModule mem_cache_module>
> CacheEnable mem /
> CacheStoreNoStore On
> MCacheSize 4096
> MCacheMaxObjectCount 200
> MCacheMinObjectSize 1
> MCacheMaxObjectSize 524288
> </IfModule>
> </IfModule>
>
> #
> # Worker MPM settings
> #
> <IfModule mpm_worker_module>
> StartServers 1
> MaxClients 250
> ThreadsPerChild 50
> MinSpareThreads 25
> MaxSpareThreads 75
> </IfModule>
>
> When I startup Apache, I notice there are multiple httpd( 1 parent and 2
> child ) instances.
> Question I had is,
>
> 1. Does this mean that at any time there are could be multiple memory
> caches equal to the number of running
> httpd instances ? or is the cache somehow shared between child processes ?
Yes, there is a separate cache for each child process.
>
> 2. How are requests handled ? For instance, do all requests go the first
> child process, till it has no
> more free threads at which point the the second child process is called
> into action ?
The allocation is essentially random depending on who is holding the
accept mutex.
You should really consider using mod_disk_cache instead. It uses a
cache that is shared between all processes, and is often as fast or
faster than mod_mem_cache because the OS can optimize the memory
caching part.
Joshua.
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: [EMAIL PROTECTED]
" from the digest: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
>
>
> Hi,
>
> I have an Apache 2.2.2 setup on a Redhat box. Caching is implemented using
> mod_mem_cache.
> Apache was built to use worker MPM. Here are the relevant sections from
> httpd.conf
>
> #
> # mod_cache/mod_mem_cache settings
> #
> <IfModule cache_module>
> <IfModule mem_cache_module>
> CacheEnable mem /
> CacheStoreNoStore On
> MCacheSize 4096
> MCacheMaxObjectCount 200
> MCacheMinObjectSize 1
> MCacheMaxObjectSize 524288
> </IfModule>
> </IfModule>
>
> #
> # Worker MPM settings
> #
> <IfModule mpm_worker_module>
> StartServers 1
> MaxClients 250
> ThreadsPerChild 50
> MinSpareThreads 25
> MaxSpareThreads 75
> </IfModule>
>
> When I startup Apache, I notice there are multiple httpd( 1 parent and 2
> child ) instances.
> Question I had is,
>
> 1. Does this mean that at any time there are could be multiple memory
> caches equal to the number of running
> httpd instances ? or is the cache somehow shared between child processes ?
Yes, there is a separate cache for each child process.
>
> 2. How are requests handled ? For instance, do all requests go the first
> child process, till it has no
> more free threads at which point the the second child process is called
> into action ?
The allocation is essentially random depending on who is holding the
accept mutex.
You should really consider using mod_disk_cache instead. It uses a
cache that is shared between all processes, and is often as fast or
faster than mod_mem_cache because the OS can optimize the memory
caching part.
Joshua.
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: [EMAIL PROTECTED]
" from the digest: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
