Hi Andreas,
I'm the original author of the chance feature and indeed, there might be
some trouble using it in a receive (something I warned for when
submitting, interacting in code is not always that evident).
At the other hand, the use on NOP, SEND and PAUSE is certainly ok (the
cases I'm using). So, if you put of NOP after the receive to choose, it
should be ok (so I had no real urgent reason to go for the last bit).
The working of chance=0.0 or 1.0 is programmed to be what you expect,
never or always.
I will review the receive case some day (when having more time), but I'm
still waiting for one of my other contributions (correction on child
handling) to be integrated: it's a little difficult to have a lot of
changes hanging around without being integrated, it all gives different
versions that need verification or changes when a new sipp is
available).
Best regards,
MarcVD
--------------------------- original question
-----------------------------------------------
Message: 1
Date: Fri, 8 Jun 2007 10:47:32 +0200
From: Andreas Bystr?m <[EMAIL PROTECTED]>
Subject: [Sipp-users] Question on random branching
To: <[email protected]>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="iso-8859-1"
Hi all,
I have been browsing through the mail archive but cant find any answer
to the problem I have. I hope that someone can help me
I'm writing a AUS scenario and it starts with receiving an Invite. Then
I want that in 99% of the cases the call should be setup with 180, 200
and so on. However, for 1% of the calls I want to send 500 Server Error
instead (to simulate a specific case). When reading the manual I found
the "chance"
command and tried to using that. This is how my script starts:
<recv request="INVITE" crlf="true" next="1" chance="0.01"> </recv>
Now label=1 is jumping to a place where I send 500, and if I dont jump
the script continues with setting up a simple call.
The problem is when I run this scenario Iget the 500 in all cases, not
only in 1% of the cases. Looking again in the documentaiton I can only
find examples where test and chance is used togheter. But it also says
that "test and chance can be combined, i.e to have..." so I guess I
should be able to use chance alone? If this is not possible, can I add a
test that is always true?
Thanks in advance!
// Andreas
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Sipp-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sipp-users
