[Users] choosing transport according to message size

[JF]

Answering myself...
I guess the only available way is through NAPTR lookup.

JF

On 5/11/06, JF <jfkavaka at gmail.com> wrote:
> Also, we cannot use the t_relay() function with no params if we want
> to force transport proto, we have to use
> t_relay("[proto:]host[:port]"), and specify a host in the script. Am I
> right? Is there another way to force the proto?
>
> Thanks in advance.
>
> JF
>
> On 5/11/06, JF <jfkavaka at gmail.com> wrote:
> > Thanks for your answer.
> >
> > This is a real problem when there are several SIP hops adding Vias and
> > Record-Routes, especially with INVITEs carrying SDPs as large as 700
> > bytes... Messages easily grow bigger than the MTU.
> > Is there a recommended way of handling this using openser? Anyone else
> > facing this kind of problems? The safest way is to use always TCP...
> >
> > JF
> >
> > On 5/10/06, Bogdan-Andrei Iancu <bogdan at voice-system.ro> wrote:
> > > Hi,
> > >
> > > no, there is no way to figure out the size of the outgoing package. The
> > > only automatic protocol selection is via NSPTR records (if the target
> > > has such records) - please see the SIP location for this.
> > >
> > > also you can choose the proto via the script.
> > >
> > > regards,
> > > bogdan
> > >
> > > JF wrote:
> > >
> > > > Hi all,
> > > >
> > > > Is there any function in openser that selects the transport protocol
> > > > (TCP or UDP) according to the message size? E.g. if the message going
> > > > out is bigger than MTU, use TCP... or does the script writer have to
> > > > take this into account every time a relaying function is called?
> > > > Also, in this second case, is the actual outgoing message size
> > > > available in the script (i.e. after lumps are applied)?
> > > >
> > > > Thanks in advance,
> > > >
> > > > JF
> > > >
> > > > _______________________________________________
> > > > Users mailing list
> > > > Users at openser.org
> > > > http://openser.org/cgi-bin/mailman/listinfo/users
> > > >
> > > >
> > >
> > >
> >
>