You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've got a failure testcase with an UAS that receives a call via REFER, and returns bad SDP in the 2xx.
It works fine in UDP, and intermittently fine over TCP, because when the preceeding system receives the 2xx, it responds with ACK and BYE really fast, which sometimes ends up in the same TCP segment/ethernet frame.
In those cases, SIPp will fail to properly parse both the ACK and the BYE, so the BYE never gets a 200 OK.
And yes, I am aware that there are 2 BYEs being sent by the preceeding hop, caused by my PjSIP stacl behaving (imho) a bit strange on SDP errors, but since this testcase works fine on UDP, the error is still in the TCP framer/parser I feel.
The text was updated successfully, but these errors were encountered:
We've been struggling with reliability of our TCP SIPp tests for exactly this reason for a few years now. It seems that SIPp is able to assemble the SIP message from 2 TCP frames but is unable to handle a frame with more than one SIP message.
I've got a failure testcase with an UAS that receives a call via REFER, and returns bad SDP in the 2xx.
It works fine in UDP, and intermittently fine over TCP, because when the preceeding system receives the 2xx, it responds with ACK and BYE really fast, which sometimes ends up in the same TCP segment/ethernet frame.
In those cases, SIPp will fail to properly parse both the ACK and the BYE, so the BYE never gets a 200 OK.
SippFailsTcpSipParsing.pcapng.gz
And yes, I am aware that there are 2 BYEs being sent by the preceeding hop, caused by my PjSIP stacl behaving (imho) a bit strange on SDP errors, but since this testcase works fine on UDP, the error is still in the TCP framer/parser I feel.
The text was updated successfully, but these errors were encountered: