[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sip-implementors
Subject:    Re: [Sip-implementors] Constructing an ACK request (non 2XX based)in a stateless case
From:       "Teodor Georgiev" <teodor () visp ! net ! lb>
Date:       2009-09-29 23:20:54
Message-ID: 0B4938D03A6C437AAFF7A7573A76882F () md2kbook
[Download RAW message or body]



Hi Brett and thanks for the tips.

1. That is the way I have done it - two times a CR\LF

2. yes, Max-Forwards is a must, first I put it, then I removed it - doesn't 
help in both cases.

3. The system I have made runs in bridge mode, it doesn't have an IP, so to 
say. It transparently rewrites the SIP packets, so 100% I am not changing 
the IP or
sending from a different one.

----- Original Message ----- 
From: "Brett Tate" <brett@broadsoft.com>
To: "Teodor Georgiev" <teodor@visp.net.lb>; 
<sip-implementors@lists.cs.columbia.edu>
Sent: Tuesday, September 29, 2009 9:17 PM
Subject: RE: [Sip-implementors] Constructing an ACK request (non 2XX 
based)in a stateless case


>> What kind of darn errors am I making?
>
> Here are a few things to check:
>
> 1) Be sure your ACK request completes with a 2 carriage-return line-feeds 
> (1 to complete the last header and another completing the ACK (not 
> containing a body).
>
> 2) The ACK is missing a Max-Forwards header; rfc3261 section 20.1 table 1 
> indicates that the header is mandatory.  However I doubt that this is 
> really the issue.
>
> 3) You mentioned a stateless proxy; however the number of via 
> headers/entries does not reflect the involvement a proxy.  I mainly 
> mention it in case your "proxy" is sending the ACK from a different IP 
> address.  Depending upon a vendor's interpretation of rfc3261 section 
> 17.2.3 and services such as using Access Control Lists, it can cause a 
> problem.
>
> 


[prev in list] [next in list] [prev in thread] [next in thread]