If your configuration has remained the same and you aren't seeing any
problems in the logs, then I suspect that all that's happened is that your
users are being a bit more adventurous than before. Caches remove
redundancy from a given traffic stream, and if your users are doing a lot
of random "surfing" then there won't be many requests for the same
objects.
BTW, upgrading to the 2.0.30+ kernel can improve your cache's performance
quite a bit. If you go to 2.0.33 make sure to use the patch that Dancer
posted here a day or two ago.
-Bill
On Sat, 11 Apr 1998, Graham Somers wrote:
>
>
> Hi
>
> I recently successfully upgraded our cache from squid 1.1.17 to 1.1.20+retry
> keeping the same cache data. The cache hits for our proxy remained at a 3 month average of 37.9%.
>
> We have just upgraded the machine's storage devices and restarted the cache on a new 5G scsi device. The cache hits are now sitting at about 31%. We have been running for at least a comparable length of time and there is more data in the cache than before.
> The squid.conf has not changed other than the larger cache size and proportionally more memory.
> We are not running out of file descriptors and there does not seem to be anything untoward in the cache.log
>
> The machine is running Linux 2.0.29 (Slackware)
>
> Is squid 1.1.20+retry less efficient than the older versions or am I missing something in the configuration.
>
> Any ideas?
>
> Graham
>
>
Received on Sat Apr 11 1998 - 22:24:00 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:39:39 MST