I checked the modifications in.
From and To tags are now:
tag=[pid]SIPpTag00[call_number], with "00" being the scenario number.
In addition, I changed as you suggested scenarios 7 and 8 to have different tags for REGISTER and INVITE.
Let me know if this looks ok to you.
SVN 60.
Olivier.
On 10/16/06, Peter Higginson <
[EMAIL PROTECTED]> wrote:
Olivier,
I like that suggestion. I would go further and make the UAS ones different
from the UAC (even though the pid will change) or even all the examples
different so people get the idea they should vary it.
An important thing is in 7 and 8 to use a different tag on the REGISTER from
the INVITE and also in the replies. They never should be the same.
Regards,
Peter
Peter Higginson
Newport Networks Ltd,
Direct line 01494 470694
http://www.newport-networks.com/
-----Original Message-----
From: Olivier Jacques [mailto:[EMAIL PROTECTED]]
Sent: 16 October 2006 08:48
To: Peter Higginson
Cc: 'Jeff Ollier'; [email protected]
Subject: Re: [Sipp-users] Registration and calls from multiple users
Peter Higginson wrote:
>
> You will need a number of SIPP machines (or at least processes - I use
> Linux myself on which several processes are easy to do).
>
> It depends quite what your other equipment is. Assuming it is some
> form of CA then what I would do for 100 calls is have three processes,
> and have "-m 50" in the calling parameters of (a) and (c):
>
> a) do 50 registers to register some users to receive the calls (first,
> before (b) and (c))
>
> b) start a server to receive the calls (a) registered for
>
> c) start a client to register and make a call for each of 50 users
>
> If you are testing some form of proxy then you can write the server to
> do the registration and just use (b) and (c) but it does not sound
> like you are doing that.
>
> In your register, you need to vary the ua name in the To: From: and
> Contact:. I would use 410x[call_number] or similar where x is
> different between (a) and (c).
>
> Also "tag=[call_number]" is a bit simplistic. Make sure that your
> invite and the response have different tags from each other and the
> registers. "tag=[pid]tkhyexy[call_number]" would be more realistic
> where you vary the bit in the middle whenever the protocols allows.
> (The [pid] means if you run the test again you get a different value.)
> If it is a good RFC 3261 implementation the tag is as important as the
> Call-ID.
>
> Peter
>
Peter and all,
I think we should change the default SIPp scenarios so that the tag is a
bit more randomized. What about "tag=[pid]SIPpTag[call_number]"? Any
other suggestion?
Olivier.
---------------
This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorized copying, disclosure or distribution of the contents in this e-mail is strictly forbidden.
---------------
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Sipp-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sipp-users
--
HP OpenCall Software
http://www.hp.com/go/opencall/
------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________ Sipp-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sipp-users
