Today I examined the latency (service time) for requests that caused a cache hit and a cache miss. Both were taking the same time. That didn't sound to good.
So I umounted the cache directory, created a new filesystem and restarted squid. After that, the load dropped considerably (from 3 to 1), mostly due to less IO wait. Cache hits were being served faster than cache misses. So all went back to normal. So, can a cache be "too big"? Current setup (well, the relevant bits at least): maximum_object_size 50 MB cache_dir diskd /squid-cache 45000 16 16 request_header_max_size 15 KB request_body_max_size 750 MB The machine is 32 bits, MemTotal: 2060960 kB -- Ralf Hildebrandt Geschäftsbereich IT | Abteilung Netzwerk Charité - Universitätsmedizin Berlin Campus Benjamin Franklin Hindenburgdamm 30 | D-12203 Berlin Tel. +49 30 450 570 155 | Fax: +49 30 450 570 962 [email protected] | http://www.charite.de
