>I'd also check "df -i", maybe you're running out of inodes in your cache dir Doesn’t seem so.... df -i Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on /dev/ad0s1a 81029320 28328770 46218206 38% 613110 9867528 6% / devfs 1 1 0 100% 0 0 100% /dev /dev/ad0s1d 30462636 3318960 24706666 12% 8929 3947805 0% /var /dev/ad4s1d 18933862 18775082 158780 99% 426751 2022655 17% /mnt/cache1 /dev/ad8s1d 18935374 8194906 10740468 43% 263890 2185516 11% /mnt/cache2 devfs 1 1 0 100% 0 0 100% /var/named/dev On 11/1/08, Amos Jeffries <squid3@xxxxxxxxxxxxx> wrote: > Marcel Grandemange wrote: >> Good day users. >> >> >> I seem to have a performance issue where my squid server doesn't seem to >> exceed 400k on objects in cache, it is not the specs of the box as im able >> to with >> Different proxy software achieve 8m on a P3. >> >> Advise? Need More info? >> > > Yes, > * version of squid (including release number)? > * some config info. > > Specific to your problem some things to check are > is 400k mean 400k objects cached or 400kB/sec fetch speeds? > delay pools in use? > single cache_dir per disk spindle? > > we may also need to check efficient use of access controls. Some types > like regex are known to cause major speed bumps. > > Amos > -- > Please be using > Current Stable Squid 2.7.STABLE5 or 3.0.STABLE10 > Current Beta Squid 3.1.0.1 >
