search

jinzigui / imsdroid

Automatically exported from code.google.com/p/imsdroid
0 stars 0 forks source link

Bad subsequent unpublish (RFC 3903) #63

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1.Start IMSDroid and activate presence
2.Change presence text
3.Sign out

What is the expected output? What do you see instead?

A unpublish request (PUBLISH expires 0) must not be done in a dialog since
"A PUBLISH request does not establish a dialog" according to RFC 3903
But the client makes its PUBLISH expires 0 request as a subsequent of the 
previous one, it should do it in a new transaction

What version of the product are you using? On what operating system?
imsdroid-1.0.285

Original issue reported on code.google.com by cedric.d...@gmail.com on 17 Sep 2010 at 1:13

Attachments:

GoogleCodeExporter commented 9 years ago

Original comment by boss...@yahoo.fr on 21 Sep 2010 at 5:21

GoogleCodeExporter commented 9 years ago 
Hi,

I'm a little confused with your last comment ;)

Do you need more informations?

Regards

Original comment by cedric.d...@gmail.com on 21 Sep 2010 at 7:59

GoogleCodeExporter commented 9 years ago 
It's normal as there is no comment. I just changed the status from "New" to 
"accepted".
I'll make the changes for the next release. The UnPublish request will have the 
same call-id but with an empty to-tag.

Original comment by boss...@yahoo.fr on 21 Sep 2010 at 8:09

GoogleCodeExporter commented 9 years ago 
1.0.292

Original comment by boss...@yahoo.fr on 28 Sep 2010 at 9:29

GoogleCodeExporter commented 9 years ago 
Hi,

Thanks for the fix

But it seems that doubango revision 487 does not reflect it, do you plan to 
make the fix also in this project?

Regards

Original comment by cedric.d...@gmail.com on 1 Oct 2010 at 11:28

GoogleCodeExporter commented 9 years ago 
Could you please send Wireshark trace with both the initial PUBLISH and 
unPUBLISH/rePUBLISH?

Original comment by boss...@yahoo.fr on 4 Oct 2010 at 2:37

GoogleCodeExporter commented 9 years ago

Original comment by boss...@yahoo.fr on 4 Oct 2010 at 2:38

GoogleCodeExporter commented 9 years ago 
Hi,

Here is a wireshark trace made with imsdroid r294 but with libtinyWRAP.so built 
from 
doubango r487

Unlike the embedded libtinyWRAP.so in imsdroid r294, you can see that the 
unpublish request (trame 4) still contains the To tag paremeter and badly make 
this request as subsequent.

Regards

Original comment by cedric.d...@gmail.com on 4 Oct 2010 at 3:03

Attachments:

GoogleCodeExporter commented 9 years ago 
Very strange, I cannot reproduce the issue.
According the trace (User-Agent header) you have modified the source code which 
means that you can debug the code. Is it rigth?
Could you please see if "tsip_dialog.c" line 683 is called for the PUBLISH 
request? On line 682 we check that the response is neither REGISTER nor PUBLISH 
before updating the to-tag.

Original comment by boss...@yahoo.fr on 5 Oct 2010 at 3:46

GoogleCodeExporter commented 9 years ago 
Hi, sorry for the delay and sorry for the mistake, you're right, my private 
node was not well merged.

Thanks again for your support

Regards

Original comment by cedric.d...@gmail.com on 6 Oct 2010 at 1:51

GoogleCodeExporter commented 9 years ago 
You are welcome.
I will mark the issue as fixed. If you find another pb related to this issue I 
can reopen it.

Original comment by boss...@yahoo.fr on 6 Oct 2010 at 2:27