Re: [squid-users] Squid on DualxQuad Core 8GB Rams - Optimization - Performance - Large Scale - IP Spoofing

> Dear Amos,
> Thank you for your reply, check my comments:
>
> Amos Jeffries wrote:
>> Haytham KHOUJA (devnull) wrote:

<snip>
>>>
>>> cache_mem 600 MB
>>> :: A bit confused about this, When i go higher than 2GB, Squid kills
>>> with a "out of memory" error. I have 8GB and want to maximize the use
>>> of it.
>>>
>>> cache_effective_user nobody
>>> cache_effective_group nogroup
>>> :: Security and bla bla
>>
> So i can leave it to 2GB maximum? The rest of the OS will have the rest
> of the RAM for OS purposes.

I was talking below about the cache_effective_* above. I know very little
about the memory limits in squid, not having tested any yet.

>> This is the default UID. If this is going to be a standard config
>> these MUST not be explicitly set.
>> Also when GID is configured as above, will in fact cause a
>> squid-specific deviation from the configured OS-level security policy.
>>
>> They are no longer to be used, unless the machine-specific setup
>> requires it AND the admin knows how to setup for them properly.
>>
>>> cache_replacement_policy heap LFUDA
>>> memory_replacement_policy heap GDSF
>>> :: Very objective, you can google about them
>>>
>>> cache_dir diskd /CACHE1 61440 16 256 Q1=144 Q2=128
>>> cache_dir diskd /CACHE2 61440 16 256 Q1=144 Q2=128
>>> cache_dir diskd /CACHE3 61440 16 256 Q1=144 Q2=128
>>> :: DISKD configuration, i'm only using 60GB of each disk
>>>
>>> cache_access_log /var/log/squid/access.log
>>
>> Obsolete option. Use access_log with same parameters instead.
>>
> Which is obsolete?

cache_access_log, it is deprecated from 2.6 and not guaranteed to be
around much longer.

Amos
Received on Sun Oct 14 2007 - 14:05:36 MDT