Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Hi!
It should work with clients using not only STUN, but the whole ICE algorithm.
btw: I analyzed google talk and they use somthing equal. The callee detects if the caller behind the same NAT (same public IP) and then they agree on local IP:port.
regards klaus
Federico Giannici wrote:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
As Klaus already mentioned, with clients which support ICE this should be no problem any more. If the clients do not support ICE, you could take a look on the MPO algorithm from Jasomi.
Regards Nils
On Thursday 25 August 2005 16:54, Federico Giannici wrote:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Nils Ohlmeier wrote:
As Klaus already mentioned, with clients which support ICE this should be no problem any more. If the clients do not support ICE, you could take a look on the MPO algorithm from Jasomi.
What is the "MPO algorithm from Jasomi"? I couldn't find it on google (actually I found same MPO algorithms, but none related to the network)... Can you give me a link (if it wasn't a joke)?
Thanks.
On Thursday 25 August 2005 16:54, Federico Giannici wrote:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
What you need is a session border control. Try this link from Jasomi : http://www.jasomi.com/ppfes.html or this from Kagoor ( now Juniper ) http://www.juniper.net/customers/support/products/vf1000.jsp
Federico Giannici ha scritto:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
beg my pardon, but I don't really think that one needs a session border control, particularly becasue of the reasons related to bad scalability, bandwidth waste and latency. Sinle point of failure is another concern.
-jiri
At 02:20 PM 8/26/2005, Carlo wrote:
What you need is a session border control. Try this link from Jasomi : http://www.jasomi.com/ppfes.html or this from Kagoor ( now Juniper ) http://www.juniper.net/customers/support/products/vf1000.jsp
Federico Giannici ha scritto:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
-- Jiri Kuthan http://iptel.org/~jiri/
Carlo wrote:
maybe you are right but I can see that FWD is using Jasomi SBC to solve nat problems.
There are also other SIP provider than FWD ;-)
klaus
http://lite.fwdnet.net/index.php?section_id=78
Carlo
Jiri Kuthan ha scritto:
beg my pardon, but I don't really think that one needs a session border control, particularly becasue of the reasons related to bad scalability, bandwidth waste and latency. Sinle point of failure is another concern.
-jiri
At 02:20 PM 8/26/2005, Carlo wrote:
What you need is a session border control. Try this link from Jasomi : http://www.jasomi.com/ppfes.html or this from Kagoor ( now Juniper ) http://www.juniper.net/customers/support/products/vf1000.jsp
Federico Giannici ha scritto:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
-- Jiri Kuthan http://iptel.org/~jiri/
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
Also, many NATs can handle hairpin media quite well. I think there is an RFC with a list of tested NATs. Another suggested solution is to test caller's and callee's public IPs and not force proxy if they match (you need to know that the internal NAT is routable and non-NATed for this to work). g-)
----- Original Message ----- From: "Klaus Darilion" klaus.mailinglists@pernau.at To: "Carlo" c.maggiolini@elitel.it Cc: "Jiri Kuthan" jiri@iptel.org; serusers@lists.iptel.org Sent: Friday, August 26, 2005 07:14 PM Subject: Re: [Serusers] Hairpin solutions?
Carlo wrote:
maybe you are right but I can see that FWD is using Jasomi SBC to solve nat problems.
There are also other SIP provider than FWD ;-)
klaus
http://lite.fwdnet.net/index.php?section_id=78
Carlo
Jiri Kuthan ha scritto:
beg my pardon, but I don't really think that one needs a session border control, particularly becasue of the reasons related to bad scalability, bandwidth waste and latency. Sinle point of failure is another concern.
-jiri
At 02:20 PM 8/26/2005, Carlo wrote:
What you need is a session border control. Try this link from Jasomi : http://www.jasomi.com/ppfes.html or this from Kagoor ( now Juniper ) http://www.juniper.net/customers/support/products/vf1000.jsp
Federico Giannici ha scritto:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
-- Jiri Kuthan http://iptel.org/~jiri/
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
Jiri Kuthan wrote:
beg my pardon, but I don't really think that one needs a session border control, particularly becasue of the reasons related to bad scalability, bandwidth waste and latency. Sinle point of failure is another concern.
Yes, I wasn't looking for "far end" (in the service provider) solutions, but the "near end" seems appropriate...
Anyway, I was looking for a smaller solution, something for small offices, from 2 to 10-20 phones.
Now I was asking myself if an Asterisk solution could work. Can Asterisk work as a SIP proxy behind a NAT? Does it have a STUN client?
Thanks.
At 02:20 PM 8/26/2005, Carlo wrote:
What you need is a session border control. Try this link from Jasomi : http://www.jasomi.com/ppfes.html or this from Kagoor ( now Juniper ) http://www.juniper.net/customers/support/products/vf1000.jsp
Federico Giannici ha scritto:
Considering that we DON'T want to use an RTP Proxy (bad scalability, waste of bandwidth and longer delays), is there any solution to the "hairpin" problems?
Are there any "boxes" that look at the SIP messages and short-circuit calls from and to the local network?
Thanks.
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
-- Jiri Kuthan http://iptel.org/~jiri/
Serusers mailing list serusers@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serusers
On 8/26/05, Federico Giannici giannici@neomedia.it wrote:
beg my pardon, but I don't really think that one needs a session border control, particularly becasue of the reasons related to bad scalability, bandwidth waste and latency. Sinle point of failure is another concern.
Yes, I wasn't looking for "far end" (in the service provider) solutions, but the "near end" seems appropriate...
Anyway, I was looking for a smaller solution, something for small offices, from 2 to 10-20 phones.
Have a look at the Intertex SurfinBird IX67, I believe that it will do what you need and is fairly low cost.
A.
-
Adam Sherman -
Carlo -
Federico Giannici -
Greger V. Teigre -
Jiri Kuthan -
Klaus Darilion -
Nils Ohlmeier