On Sun, 25 Oct 1998, Parabola wrote:
> I discovered that if I set the proxy of GetRight to Squid 2.0 Patch 2
> running at my Linux box, every time GetRight tries to resume the download
> of a file (say, 10Mb was downloaded last time and I want to start from
> there), Squid actually tries to pull the _whole_ thing back, ie Squid will
> not respond to the request of GetRight until it finishes download the first
> 10Mb!
First of all, check you quick_abort and other settings to be sure that the
file was in fact successfully downloaded the first time. Try to send an
only-if-cached HEAD request to Squid to see if the file is there.
If the file is indeed in the cache, check the HTTP headers that GetRight
sends to Squid. If ranges in Range header are out of order, you are out of
luck. If they are in order, then either the object was not cachable from the
start, or the object was stale during the second request, or there is a bug.
Since range request support was added recently, there may be cases we do not
cover yet. If you believe there is a bug in how we handle _in order_ ranges,
please send us some debugging to illustrate your point.
> So is there any way to get around it? I mean, I don't care if the big file
> would get cached or not, but at least byte-range request should be allowed
> to get through _without_ retrieving the whole file.
As a temporary solution, you can set we_do_ranges variable in http.c to 0
always. Then, I think, Squid will forward _all_ range requests to the origin
server as it used to do. Perhaps we should have a yet another acl to filter
objects that should not be prefetched on range requests. :(
Alex.
Received on Sun Oct 25 1998 - 07:35:58 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:42:46 MST