Bugs item #1930055, was opened at 2008-03-31 12:47
Message generated for change (Comment added) made by gdesmott
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=756076&aid=1930055&group_id=143636

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Pending
Resolution: Fixed
Priority: 5
Private: No
Submitted By: Mikhail Zabaluev (mzabaluev)
Assigned to: Pekka Pessi (ppessi)
Summary: Unregister when a new public binding is detected

Initial Comment:
When outbound detects a new address binding, any successful registration by the 
same NUA handle should be explicitly unregistered. This may be needed in two 
cases:
1) weird configurations with NAT and no authentication;
2) when the binding changes on a NAT middle box.

See also the discussion on a maemo bug:
https://bugs.maemo.org/show_bug.cgi?id=3056

----------------------------------------------------------------------

Comment By: Guillaume Desmottes (gdesmott)
Date: 2008-12-04 22:45

Message:
I can confirm this bug. I'm unable to login to ekiga.net using
telepathy-sofiasip git master.

----------------------------------------------------------------------

Comment By: Pekka Pessi (ppessi)
Date: 2008-11-26 20:32

Message:
Mikhaila's patch is available in darcs. 

----------------------------------------------------------------------

Comment By: Johan Brannlund (jbrnd)
Date: 2008-11-16 20:49

Message:
In the thread
http://comments.gmane.org/gmane.comp.freedesktop.telepathy/2115 , Mikhail
Zabaluev said that not being able to log into ekiga.net with Empathy is due
to this bug. If that's really the case, then this bug is not fixed. I'm
running Empathy on Ubuntu 8.10 which has sofiasip 1.12.9 and I still can't
login (actually, this used to work with older versions).


----------------------------------------------------------------------

Comment By: Pekka Pessi (ppessi)
Date: 2008-11-12 18:41

Message:
Fixed in 1.12.9.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2008-10-27 19:15

Message:
There is a patch in Darcs mainline (searchable by this bug ID) that had the
fix for the problem. Unfortunately, it was an overkill, so now even those
contacts get submitted with expires=0 which were never successfully
registered. This breaks registration with proxies that dislike the
unsuccessful contact (e.g. for being in a private IP address segment), even
if it has expired. Reopening because the requirement for "successful"
registration in the bug description was not met.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2008-04-04 16:01

Message:
Logged In: YES 
user_id=313104
Originator: YES

It's been working as described, the bug is invalid.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2008-04-04 14:43

Message:
Logged In: YES 
user_id=313104
Originator: YES

> The previous contact is added to the new REGISTER request with
expires=0
parameter.

Was it implemented in Sofia-SIP for a long time?

----------------------------------------------------------------------

Comment By: Pekka Pessi (ppessi)
Date: 2008-04-04 14:37

Message:
Logged In: YES 
user_id=52043
Originator: NO

The previous contact is added to the new REGISTER request with expires=0
parameter. In other words, the REGISTER request is used to both unregister
the old contact and register the newly discovered contact.

While this is very basic REGISTER functionality from pre-RFC-2543 era, it
may confuse some proxies. Please confirm if that is so.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2008-04-02 10:59

Message:
Logged In: YES 
user_id=313104
Originator: YES

Maybe it's not a good idea to unregister _before_ the contact is updated.
But it could be done afterwards (after the whole update transaction, or
perhaps just firing an un-REGISTER request right after the REGISTER with a
contact update is sent?).
OTOH, it can bring more issues with saner proxies, than it solves with a
few not-so-sane examples.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=756076&aid=1930055&group_id=143636

------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you.  Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel

Reply via email to