Bug ID |
9007
|
Summary |
Switching SSRC values in RTP stream
|
Classification |
Unclassified
|
Product |
Wireshark
|
Version |
1.10.0
|
Hardware |
x86
|
OS |
Windows 8
|
Status |
UNCONFIRMED
|
Severity |
Major
|
Priority |
Low
|
Component |
Dissection engine (libwireshark)
|
Assignee |
[email protected]
|
Reporter |
[email protected]
|
Build Information:
--
I have a SIP with RTP capture of a call, and a media recording of the same
call. The media in one direction flows like this:
Call is answered and speech is heard, this RTP stream has an SSRC value of A1
(for example).
The RTP stream is then changed to a music on hold for 10 seconds. The music on
hold has SSRC of MOH (again simplified for sake of describing the issue).
Then, the original stream A1, is reconnected.
So we have, A1 -> MOH -> A1, and the independent media recording and packet
analysis describe this no problem.
The issue is on the RTP player for this call, on this RTP stream (which keeps
the same UDP ports and IP addresses at both side of the connection), the audio
gets reconstructed like this:
A1 -> silence -> A1 -> MOH
So rather than play back the music on hold where it actually occurred in the
call, it sticks it on the end.
I will seek permission from the parties on the call to upload the file if
needed but that about describes the issue in a nutshell.
You are receiving this mail because:
- You are watching all bug changes.