Hi,
Thanks for the prompt response.
Please see inline
>From: Henrik Nordstrom <hno@squid-cache.org>
>To: osuser os <osuser@hotmail.com>
>CC: squid-users@squid-cache.org
>Subject: Re: [squid-users] Squid 2.5 and Range requests
>Date: Sun, 25 Apr 2004 11:40:14 +0200 (CEST)
>
>On Sun, 25 Apr 2004, osuser os wrote:
>
> > But if the file size > 20 KB, I see that the full file is still being
> > retrieved, but is not being put in the cache. (future requests starting
>with
> > range 0 are resulting in full file being retrieved again and not put in
>the
> > cache).
>
>Probably aborted by quick_abort_...
>
>The default setting is to abort any downloads when the client disconnects
>if more than 16 KB remains when the client disconnected.
I see the same behavior even if the client sends "Proxy-Connection:
Keep-alive" header. Squid is also responding with "Proxy-Connection:
Keep-alive" and the connection is not aborted. But the file still doesnt get
cached.
Is there any way of modifying the default setting to abort the download?
>
> > That means that server bandwidth consumed is a lot more than client
> > bandwidth especially for large files.
>
>The abort should be more or less immediate when the client have got it's
>requested part. Are you sure it continues?
I confirmed from network monitor that the 1st connection is not getting
aborted immediately after getting a response.. But when I send the 2nd
request, it is not being sent on the same socket (old socket is being closed
and new socket is established). Probably a new connection is being used
because of HTTP/1.0 response. The requests are being sent using HTTP/1.1
Regards
Marc
>
>Regards
>Henrik
>
_________________________________________________________________
Watch LIVE baseball games on your computer with MLB.TV, included with MSN
Premium!
http://join.msn.com/?page=features/mlb&pgmarket=en-us/go/onm00200439ave/direct/01/
Received on Sun Apr 25 2004 - 14:43:10 MDT
This archive was generated by hypermail pre-2.1.9 : Fri Apr 30 2004 - 12:00:02 MDT