Re: caching mirrored files more efficiently

Laszlo Vecsey wrote:
>
> I dont know if squid already does this, if it does thats great. But a lot
> of sites have large downloads that are mirrored to multiple sites (tucows,
> pointcast, netscape are a few big ones to name a few), and if the users
> dont click on the same mirror site an unnecessary download will take
> place. So it would be nice if squid could notice that 1) the file name and
> file size are identical, and 2) maybe gain a little more certainty by
> checking some of the paths, to see if they look similar. This would rule
> out most of the cases where a file might be the same size name etc but
> have different binary contents. Perhaps the timestamp would make this
> possiblity even more remote.

Squid doesn't do this. Some of us have redirectors rewriting requests
for certain files/paths to our local mega-mirrors to consolidate
caching.

D

-- 
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GAT d- s++: a C++++$ UL++++B+++S+++C++H++U++V+++$ P+++$ L+++ E-
W+++(--)$ N++ w++$>--- t+ 5++ X+() R+ tv b++++ DI+++ e- h-@ 
------END GEEK CODE BLOCK------