[SFTF] Re: SFTF-0.9.2 exception
Hi Jonathan,
cc'ed the mailing list as the answer might be intersting for others as well.
On Wednesday 11 August 2004 22:22, Jonathan Story wrote:
> Thanks to you, the non-interactive tests completed.
>
> I happen to have subversion lying around (unused up to now) and got the
> updates, but I wish there was an explicit example "checkout" statement
> on the web page for dummies like me.
> (E.g. "svn co http://scm.sipfoundry.org/rep/sftf sftf".)
It was there, but a little bit hidden in the text. I moved it to a hopefully
better place.
> This is a fine set of tests. I am wondering whether cases can be created
> not just to test strange environments, but to actually mimic particular
> brand-name behaviours for interoperability testing. If so, some pointers
> would be appreciated.
You can send almost any SIP request you want. So could simulate the outgoing
and incoming request. But you can not simulate the behavior of the stack of
others implementations. Thus the simulation would not be perfect.
If you are interested in testing interoperability with other brand-name
vendors I suggest that you attend the next SIPit.
> I notice that there is a call for Audio/RTP parsing. Has anyone stepped
> up, yet? This would be very useful.
I added recently a very basic SDP parser. But to be able to send and receive
audio, the problem of listening to multiple sockets at the same time has to
be solved first. I am currently not sure if it would be better to audio in
the same process, or start a new thread for each RTP stream.
But in general I will not put too many effort into RTP support, because the
primary target of this tool is SIP testing and not audio tests (there are
other, probably commercial, tools available for that purpose).
Greetings
Nils
> Regards,
>
> Jonathan Story
>
> Nils Ohlmeier wrote:
> >Hi Jonathan,
> >
> >thanks again :-)
> >Fixed file is attached again. Are you able to update from the subversion
> >repository? Because I also fixed three other test which can fail in the
> > same situation.
> >
> >I dont care about getting bigger attachments. If would like to reduce the
> > file size, just copy the last test run from the debug.log. Mostly it is
> > only important what was received during the start of the test until the
> > failure happend.
> >
> >The exception is this case was caused by the REGISTER which was received
> >before a reply for the INVITE request. So it does not depend on running
> > the test stand alone or in a row, but on what the other side sent to the
> > test tool.
> >
> >Greetings
> > Nils
> >
> >On Wednesday 11 August 2004 19:57, you wrote:
> >>Geez, you're fast. Your fix seemed to go okay, and now another cropped
> >>up, once again while using parameters "-i -r":
> >>
> >>====
> >> Caught exception during running 'Case 211cseq'
> >>Traceback (most recent call last):
> >> File "./SFTF.py", line 187, in run
> >> j.run()
> >> File "./UserAgentBasicTestSuite/case211cseq.py", line 55, in run
> >> if (repl is not None) and (repl.code > self.code):
> >>AttributeError: SipRequest instance has no attribute 'code'
> >>===
> >>
> >>I see that the size of 'debug.log' is getting larger. Oddly, when I
> >>tried to run 'case211cseq' all by itself, it ran without throwing an
> >>exception.
> >>
> >>Regards,
> >>
> >>Jonathan Story
