Caching of application/x-byterange

Greetings,

I have noticed thatmore and more servers seem to be supporting the byterange
query used by, in our case, the Adobe Reader plug-in for Netscape. I find
this most unfortunate because Squid appears incapable of caching these
byterange queries. Thus, everyone ends up going back to the source server,
which can be very expensive bandwidth wise given that PDFs are often quite
large (the ones we're looking at are often >2meg, transfered over a 56k DS0).
I have not found an option in the Acrobat plug-in to disable the use of its
streaming byterange mode. Has anyone else noticed this problem and come up
a solution for caching this stuff? I don't care if the user has to wait for
the entire PDF to download before they can see it; in the long run, caching
the entire PDF will be more benifitial to the entire user population.

Thanks,
-=> CAW

FWIW: We use Squid 1.1.18 (both VM and NOVM versions) on various platforms.
However, this has been noticed with several verions of Squid now.
Received on Wed Nov 12 1997 - 14:37:10 MST