reSIProcate

Mailing Lists

Source Code

Google Search:



< Previous by Date Date Index Next by Date >
< Previous in Thread Thread Index Next in Thread >

Re: [reSIProcate] [ReSIProcate_1.1_RC2] submit a patch forDUM/ClientPublication.cxx

So, here's a revised patch. Since 412 doesn't necessarily mean that the resource is gone, we probably should be calling onFailure. 

Index: resip/dum/ClientPublication.cxx
===================================================================
--- resip/dum/ClientPublication.cxx     (revision 7014)
+++ resip/dum/ClientPublication.cxx     (working copy)
@@ -106,10 +106,19 @@
       {
          if (code == 412)
          {
-            InfoLog(<< "SIPIfMatch failed -- republish");
-            mPublish->remove(h_SIPIfMatch);
-            update(mDocument);
-            return;
+            if(mPublish->header(h_Expires).value() != 0)
+            {
+               InfoLog(<< "SIPIfMatch failed -- republish");
+               mPublish->remove(h_SIPIfMatch);
+               update(mDocument);
+               return;
+            }
+            else
+            {
+               handler->onFailure(getHandle(), msg);
+               delete this;
+               return;
+            }
          }
          else if (code == 423) // interval too short
          {
If I see no objections today, I will be committing this evening, and RC3 will follow on Friday. 

Best regards,
Byron Campen
Okay, I'm starting to feel a little easier about this. The 412 may mean the resource is gone, but it may just mean that the resource has changed due to a PUBLISH from another UA. In either case, we probably don't need to continue trying to remove it. Does anyone have a problem with this patch (ie, can give us a reason why the patch might cause incorrect behavior)? 

Best regards,
Byron Campen


Sorry,

I made a mistake. I am saying that :
the client publishes again with expires set to 0, no SIP-If-Match and the last body sent. 

Regards

Fabrice ROUILLIER


-----Message d'origine-----
De : Byron Campen [mailto:bcampen@xxxxxxxxxxxx]
Envoyé : mardi 20 mars 2007 20:10
À : zze-Omnipresence ROUILLIER F ext RD-MAPS-REN
Cc : Michael Froman; resiprocate-devel
Objet : Re: [reSIProcate] [ReSIProcate_1.1_RC2] submit a patch forDUM/ClientPublication.cxx
I'm confused. Are you or are you not saying that the PUBLISH is going out with a body? 

Best regards,
Byron Campen


Michael,

You are right on your analysis.

But, the fact is (and I do not mentionned it before) that when the
Dum::ClientPublication removes the SIP-If-Match header and updates the 
publication, the previously sent body is also attached (see update
method). So the client publishes again with expires set to 0, no
SIP-If-Match and no body. Then the ESC responds with a 412 and so
one...
Perhaps, is there also a bug in the end() method, the document may be 
deleted ?

Regards

Fabrice ROUILLIER

-----Message d'origine-----
De : resiprocate-devel-bounces@xxxxxxxxxxxxxxxxxxxx
[mailto:resiprocate-devel-bounces@xxxxxxxxxxxxxxxxxxxx] De la part de Michael Froman Envoyé : lundi 19 mars 2007 20:04 À : resiprocate- devel 
Objet : Re: [reSIProcate] [ReSIProcate_1.1_RC2] submit a patch
forDUM/ClientPublication.cxx

Actually, I not sure that there is a bug here.

In fact publishing again with expires set to 0 and without a Sip-
if-match will raise a 412 again and again !!!
If the client publishes again with expires set to 0 (and no body) and 
no SIP-If-Match, the ESC should be responding with a 400 Invalid
Request as detailed in RFC3093, Section 6 (Processing PUBLISH
Requests), step 5:
       *  If the request has no message body and contained no
entity- tag,
          the ESC SHOULD reject the request with an appropriate
response,
such as 400 (Invalid Request), and skip the remainder of the 
          steps.  Alternatively, in case either ESC local policy or
the
          event package has defined semantics for an initial
publication
          containing no message body, the ESC MAY accept it.

What implementation is responding to the rePUBLISH with a 412?


Regards,
Michael Froman.


On Mar 16, 2007, at 2:04 PM, Byron Campen wrote:


        Well, we haven't exactly codified who is responsible for applying
patches. Usually it just goes to whoever knows the code fairly well, 
and is around. However, IETF is happening next week, so a lot of
people are in the air right now (both figuratively and literally).
DUM
is something that I have just started wading into, and I am uneasy
about applying patches without feedback from those who wrote most of 
that code.

Scott, have you looked at this?
As for when the next release is, the answer is soon (I had originally 
intended to designate 1.1-RC2 as the official release this evening,
but since a couple of bugs have been discovered in the last few days, I'll have to wait for the fixes and cut RC3, probably sometime early 
next week.)

Best regards,
Byron Campen



Byron
I do not think that a call to handler->onFailure() is necessary, the 
aim is "in fine" to do the unPublish.
Just another question, who is responsible of merging this patch into 
the reSIProcate project ?

Any idea for the next release date ?

Best Regards

Fabrice ROUILLIER

De : Byron Campen [mailto:bcampen@xxxxxxxxxxxx] Envoyé : jeudi 15
mars 2007 22:50 À : zze-Omnipresence ROUILLIER F ext RD-MAPS- REN Cc 
:
resiprocate-devel; Scott Godin Objet : Re: [reSIProcate]
[ReSIProcate_1.1_RC2] submit a patch for DUM/ClientPublication.cxx

Good find. Now, would it be necessary to call handler->onFailure
() in
this case? Is getting a 412 considered a "failure" for an unPUBLISH? 
(As far as intent goes, it seems not to me)

Best regards,
Byron Campen


Dear reSIProcate team,

I find a bug in the implementation of the "ClientPublication"
class when handling response to a 412 message received from server. 

You previously remove the "SIP-if-match" tag and republish the
document.
This SHALL NOT be done if the 412 response is received when trying 
to end the publication (Expires header set to 0)

In that case nothing more have to be done !

In fact publishing again with expires set to 0 and without a Sip-
if-match will raise a 412 again and again !!!


void ClientPublication::dispatch(const SipMessage& msg) {

...
         if (code == 412)
         {
                        // Receive a 412 while ending a
publication, nothing more to do in this case.
if(mPublish->header(h_Expires).value () != 0 
)
                        {
                                InfoLog(<< "SIPIfMatch failed --
republish");
                                mPublish->remove(h_SIPIfMatch);
                                update(mDocument);
                                return;
                        }
                        else {
                                delete this;
                                return;
                        }
         }
         else if (code == 423) // interval too short

...
}

Hope this will be corrected in next candidate release

Best Regards

Fabrice ROUILLIER

_______________________________________________
resiprocate-devel mailing list
resiprocate-devel@xxxxxxxxxxxxxxxxxxxx
https://list.resiprocate.org/mailman/listinfo/resiprocate-devel




_______________________________________________
resiprocate-devel mailing list
resiprocate-devel@xxxxxxxxxxxxxxxxxxxx
https://list.resiprocate.org/mailman/listinfo/resiprocate-devel



_______________________________________________
resiprocate-devel mailing list
resiprocate-devel@xxxxxxxxxxxxxxxxxxxx
https://list.resiprocate.org/mailman/listinfo/resiprocate-devel

_______________________________________________
resiprocate-devel mailing list
resiprocate-devel@xxxxxxxxxxxxxxxxxxxx
https://list.resiprocate.org/mailman/listinfo/resiprocate-devel




_______________________________________________
resiprocate-devel mailing list
resiprocate-devel@xxxxxxxxxxxxxxxxxxxx
https://list.resiprocate.org/mailman/listinfo/resiprocate-devel

Attachment: smime.p7s
Description: S/MIME cryptographic signature