[hipl-users] Re: Problems with SIP Communicator
- From: Miika Komu <miika.komu@xxxxxxx>
- To: hipl-users@xxxxxxxxxxxxx
- Date: Thu, 12 Feb 2009 15:00:34 +0200
Robert Moskowitz wrote:
Hi,
Running SIP communicator alpha3 nightly build 1634, on Centos 5 with
HIPL build 14 over IPv6. I have UDP port 5060 open.
Without any hipl code going, I can set up a direct SIP call between my 2
IPv6 hosts, and the call goes through rapidly.
So I am running hipd hipfw on both hosts, and hipdnsproxy on the
initiator. I start the call and nothing happens for a LONG time. Then
I get 3 'Unhandled type 33' messages, then an 'Associating DNS HIT ....
with IP .....
6 more Unhandled messages, a Hosts file or cache match message, and this
repeats and FINALLY a call dialog starts, I see the HIP base exchange in
the tcpdump window, then ESP messages and the call finally starts. The
calling agent shows the elapse on the call, but interestingly the called
agent shows 0 secs for the call.....
So something is NOT right here that it takes so long for the call. I
have more testing to do, but something holding this up.
dns proxy is rather slow in "giving up" with records that cannot be
resolved. For example, SIP communicator tries to resolve SRV records
initially, but no such things exist. The dns proxy gives up with missing
records too slowly, causing further delay for VoIP call set up.
http://hipl.infrahip.net/bugzilla/show_bug.cgi?id=745
Other related posts:
