On Sun, 2009-03-15 at 17:10 +0100, Fredrik Thulin wrote:
> Mikael Magnusson wrote:
> > I have found a problem with loop detection in Yxa.  The Request-URI used
> > when creating the loop cookie is not the original URI but the URI after
> > any translation, which is a problem.
> ...
> > 
> > My test scenario is:
> > 
> > User A calls a telephone number N which is looked up in ENUM into URI B
> > in the home domain. URI B is handled by a regexp route, and sent to
> > local user C which has a CPL script registered. This script forks into
> > multiple local users D, E, F and G in parallel.
> > 
> > The loop occurred when receiving the request for user C.
> 
> Wow, that's some test case =).
> 
> Thanks, I'll have a real close look at this. Given your previous 
> patches, I'll assume you are correct, but I have RFC5393 on my reading 
> list and the abstract says
> 
>    "It also corrects and clarifies the description of the
>     loop-detection algorithm such proxies are required to implement."
> 
> so I think reading these clarifications are in order before changing the 
> loop detection algorithm, even if the current one in YXA does not match 
> the one in RFC3261 (since there is obviously at least one known problem 
> with that algorithm).
> 
> I'll get back to you and the list when I've had a closer look at this.
> 
> /Fredrik

Another easier way to recreate the problem is using two regexps and one
account:

Regexp1: ^sip:rege...@homedomain$ sip:rege...@homedomain
Regexp2: ^sip:rege...@homedomain$ sip:u...@homedomain

There is no problem sending an INVITE to sip:rege...@homedomain. The
request is sent in a spiral one time before it reaches the user. But an
INVITE to sip:rege...@homedomain results in 482 Loop Detected.

/Mikael

_______________________________________________
Yxa-devel mailing list
Yxa-devel@lists.su.se
https://lists.su.se/mailman/listinfo/yxa-devel

Reply via email to