Re: 1.2b20-1: Reworked client body processing [patch]

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Mon, 27 Apr 1998 23:36:28 +0200

--MimeMultipartBoundary
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Duane Wessels wrote:

> Here's my thoughts:
>=20
> I do agree that the pump.c code is not very pretty, BUT I think
> it does work with HTTP POST/PUT at least. I'm not sure how it
> interacts with FTP PUT. FTP PUT is a high priority, so we need to
> get it working for squid-1.2. We seem to have finally eliminated
> all coredumps from aborted POST requests in the beta19/beta20
> versions.

My copy of beta19 was not very stable... Maybe beta20 was (never tried).

> I applied this patch and ran my simple POST test script, and
> it didn't seem to work very well:

> 1998/04/27 14:27:01| clientReadBodyHandler: end fd=3D23 size=3D95 body_=
size=3D417 in.offset=3D0 cb=3D456294 req=3D10479000
> 1998/04/27 14:27:01| clientReadRequest: FD 23: reading request...
> 1998/04/27 14:27:01| commSetSelect: FD 23 type 1
> 1998/04/27 14:27:01| parseHttpRequest: Unsupported method '=BE=EF=E6
> =FD=F9=F4=EC=E6'

Are you sure the patch was cleanly applied? With the trace given above
parseHttpRequest can't be called...

conn->body.size_left =3D=3D 417

read more data on client connection

while(conn->in.offset > 0 && conn->body.size_left =3D=3D 0) {
  ....
  http =3D parseHttpRequest(...)

Can you please send a complete copy of your client_side.c?

/Henrik

--MimeMultipartBoundary--
Received on Tue Jul 29 2003 - 13:15:48 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:11:45 MST