Hi all,
Monitoring my cache I can see that in at least *1 time in each 10
minutes the cache response time is too high* (about 15 to 30seg). When
it happens there is *high activity in the hard disks*. The normal time
response in my test is < 1s. The script I used in my workstation is like
this:
#!/bin/sh
cont=1
date
while [ "1" = "1" ] ;
do
echo Counter: $cont
/usr/bin/time -f "%e" wget -q www.microsoft.com
sleep 2
cont=$(($cont + 1))
done;
The cache is only 10% of capacity, so it's not because of object expiration.
I was using IDE hd, so I tried to use SCSI hd but the same problem persists.
The cache is configured to be used by DISKD and almost all requests is
made by using transparent proxy.
My configuration is:
Squid Cache: Version 2.5.STABLE1
Processor: Pentium 4 CPU 2.40GHz
Memory: 1GB
HD: 2 SCSI 17GB
SCSI Controler: DPT2040
Distrib: RedHat 9.0
Any help ?
Thanks a lot.
Luciano Lima
ELO Internet
Sao Luis MA - Brazil
Received on Wed Oct 01 2003 - 06:54:20 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:20:16 MST