On 22.07 07:57, Scott Phalen wrote:
> I have searched the archives and can't seem to find a solution to my issue.
> My server sits at 92% or higher for CPU utilization 24 hours a day. During
> my peek hours it is at 99% with a load of about 35 requests per second. My
> plan is to convert this to reiserfs in the next month and spread the cache
> over both disks instead of using RAID0.
good idea and should help much. However, if it is squid who eats 99% of
CPU, it is probably due to slow machine (...) or badly configured squid...
> I am using a Dell PowerEdge 650 P4 3.0Ghz processor
> 2 - 145Gig 15K RPM hard disks - striped
> 2 Gig RAM
> Linux 2.1AS
> Squid-2.5STABLE6
> Below is my squid.conf minus the ACLs and cachemgr info:
> cache_mem 128 MB
I'd increase this probably to 1GB...
> maximum_object_size 128 MB
this is imho too much for 13GB storage...
it would probably be OK with heap replacement policies like LFUDA
(i'd use that one)
> ipcache_size 0
> fqdncache_size 0
I am not sure if 0 in these cases means unlimited - check it...
-- Matus UHLAR - fantomas, [email protected] ; http://www.fantomas.sk/ Warning: I wish NOT to receive e-mail advertising to this address. Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu. Depression is merely anger without enthusiasm.Received on Tue Aug 03 2004 - 08:36:34 MDT
This archive was generated by hypermail pre-2.1.9 : Wed Sep 01 2004 - 12:00:01 MDT