[SR-Users] gruu within dialog
samuel
samu60 at gmail.com
Tue Sep 2 11:45:26 CEST 2014
It turned out to be the NAT handling process that screwed the gruu
treatment. Kamailio modified Contact from the OK (because this user is
marked as natted) and calling fix_nated_contact modified the Req-URI of
further in-dialog requests.
I have to look at the details but, using the standard config file as basic,
the NAT flags should no be marked if is_gruu is TRUE. Shall this be
included in the standard kamailio.cfg config file?
Thanks a lot for the answer!
Samuel.
On 1 September 2014 15:46, Daniel-Constantin Mierla <miconda at gmail.com>
wrote:
> Hello,
>
> the problem is the contact coming with IP address and then used in r-uri
> with IP. In a multi-domain deployment, you cannot assume what is the right
> user id (sip address) to use in case of overlapping usernames. Think about
> rather common multi-tenant scenario where the location can be partitioned
> to different servers, based on domain.
>
> AFAIK, in case GRUU is supported, the UA has to use the give GRUU URI as
> contact for further requests. Kamailio is giving the domain and the UA
> should use it as it is. So, for me it looks as an issue in the UA, unless
> there is some other proxy in the middle changing the contact.
>
> Of course, with the flexibility of kamailio you can fix it in the config,
> like:
> - if there is gr parameter to uri and the domain part is IP (see siputils
> and ipops for appropriate functions to be used), then set $rd to the domain
> of the user.
> - the domain of the user can be discovered from various sources, depending
> on local profile and signaling (e.g, From/To headers, do a sql_query() over
> subscriber table, etc.)
>
> Cheers,
> Daniel
>
>
> On 01/09/14 15:33, samuel wrote:
>
> anoyone can provide information about how lookup function treats Req-URI
> with gruu?
>
> Thanks in advance,
> Samuel.
>
>
> On 27 August 2014 09:12, samuel <samu60 at gmail.com> wrote:
>
>> Here it goes, apologies for the length:
>>
>> The registration process is done via TLS and therefore I "can not" post
>> the trace. However, the resulting data is the following:
>>
>> AOR:: sam at domain.com
>> Contact:: sip:83652074 at M.N.O.P:34120;transport=tls Q=
>> Expires:: 569
>> Callid:: iUcVvmbsda9Yu0DGUm4exTHiZYIqwgtZ
>> Cseq:: 2
>> User-agent:: Blink 0.9.1 (Linux)
>> Received:: sip:M.N.O.P:39961;transport=TLS
>> State:: CS_DIRTY
>> Flags:: 0
>> Cflag:: 64
>> Socket:: tls:X.Y.Z.W:5061
>> Methods:: 4294967295
>> Ruid:: uloc-53fc870d-1097-4
>> Instance:: <urn:uuid:d63b1c4f-d7dc-4f4e-87f1-948123266dc0>
>> Reg-Id:: 0
>> Last-Keepalive:: 1409121941
>> Last-Modified:: 1409121941
>>
>> The call trace is the following (Trying and Ringing messages removed for
>> simplicity):
>>
>> U A.B.C.D:5060 -> X.Y.Z.W:5060
>> INVITE sip:999666222 at pstn.domain.com SIP/2.0..Via: SIP/2.0/UDP
>> A.B.C.D:5060;branch=z9hG4bK222c6640..Max-Forwards: 70..From: "111222333"
>> <sip:111222333 at A.B.C.D>;tag=as1a7b4c7d..To: <
>> sip:999666222 at pstn.domain.com>..Contact: <sip:111222333 at A.B.C.D:5060>..Call-ID:
>> 59f5
>> 579c01f8039243ec830d317df994 at A.B.C.D:5060..CSeq: 102 INVITE..User-Agent:
>> IPXAdam..Date: Wed, 27 Aug 2014 06:45:54 GMT..Allow: INVITE, ACK, CANCEL,
>> OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO, PUBLISH..Supported: replaces,
>> timer..Content-Type: application/sdp..Content-Length: 311....v=0..o=root
>> 936120945 936120945 IN IP4 A.B.C.D..s=Asterisk PBX 11.6-cert2..c=IN IP4
>> A.B.C.D..t=0 0..m=audio 12018 RTP/AVP 8 3 0 101..a=rtpmap:8
>> PCMA/8000..a=rtpmap:3 GSM/8000..a=rtpmap:0 PCMU/8000..a=rtpmap:101
>> telephone-event/8000..a=fmtp:101 0-16..a=silenceSupp:off - - -
>> -..a=ptime:20..a=sendrecv..
>>
>>
>> U X.Y.Z.W:5060 -> A.B.C.D:5060
>> SIP/2.0 200 OK..Via: SIP/2.0/UDP
>> A.B.C.D:5060;rport=5060;branch=z9hG4bK222c6640..Record-Route:
>> <sip:X.Y.Z.W:5061;transport=tls;lr;r2=on;fdrrm=82.63f;nat=yes>..Record-Route:
>> <sip:X.Y.Z.W;lr;r2=on;fdrrm=82.63f;nat=yes>..Call-ID:
>> 59f5579c01f8039243ec830d317df994 at A.B.C.D:5060..From: "111222333"
>> <sip:111222333 at A.B.C.D>;tag=as1a7b4c7d..To: <
>> sip:999666222 at pstn.domain.com>;tag=GcH-CAWXaNVzm0W314zxJF518oM-Okco..CSeq:
>> 102 INVITE..Server: Blink 0.9.1 (Linux)..Allow: SUBSCRIBE, NOTIFY, PRACK,
>> INVITE, ACK, BYE, CANCEL, UPDATE, MESSAGE, REFER..Contact:
>> <sip:sam at M.N.O.P:39961;transport=tls;gr=urn:uuid:d63b1c4f-d7dc-4f4e-87f1-948123266dc0>..Supported:
>> 100rel, replaces, norefersub, gruu..Content-Type:
>> application/sdp..Content-Length: 236....v=0..o=- 3618110757 3618110758
>> IN IP4 M.N.O.P..s=Blink 0.9.1 (Linux)..t=0 0..m=audio 50002 RTP/AVP 8
>> 101..c=IN IP4 M.N.O.P..a=
>> rtcp:50003..a=rtpmap:8 PCMA/8000..a=rtpmap:101
>> telephone-event/8000..a=fmtp:101 0-15..a=sendrecv..
>>
>> U A.B.C.D:5060 -> X.Y.Z.W:5060
>> ACK
>> sip:sam at M.N.O.P:39961;transport=tls;gr=urn:uuid:d63b1c4f-d7dc-4f4e-87f1-948123266dc0
>> SIP/2.0..Via: SIP/2.0/UDP A.B.C.D:5060;branch=z9hG4bK22a00025..Route:
>> <sip:X.Y.Z.W;lr;r2=on;fdrrm=82.63f;nat=yes>,
>> <sip:X.Y.Z.W:5061;transport=tls;lr;r2=on;fdrrm=82.63f;nat=yes>..Max-Forwards:
>> 70..
>> From: "111222333" <sip:111222333 at A.B.C.D>;tag=as1a7b4c7d..To: <
>> sip:999666222 at pstn.domain.com>;tag=GcH-CAWXaNVzm0W314zxJF518oM-Okco..Contact:
>> <sip:111222333 at A.B.C.D:5060>..Call-ID:
>> 59f5579c01f8039243ec830d317df994 at A.B.C.D:5060..CSeq: 102
>> ACK..User-Agent: IPXAdam..Content-Length:0....
>>
>> What I was refering to is that in the logs the lookup process is using
>> sip:sam at M.N.O.P, which is not found because what exists in the registrar
>> database is sam at domain.com. In the Contact header of the 200 OK the
>> local IP is used instead of the FQDN form. I might have been misleaded by
>> the logs or the gruu lookup process, but in the following lines of the code
>> (you were right about the lines and verion):
>>
>> The first log ouput comes from the following lines of lookup.c:
>>
>> 120 if(puri.gr_val.len>0) {
>> 121 /* pub-gruu */
>> 122 inst = puri.gr_val;
>> 123 LM_DBG("looking up pub gruu [%.*s]\n",
>> inst.len, inst.s);
>>
>> But afterwards, there are these lines, with the return -1 statement:
>> 154 /* aor or pub-gruu lookup */
>> 155 ul.lock_udomain(_d, &aor);
>> 156 res = ul.get_urecord(_d, &aor, &r);
>> 157 if (res > 0) {
>> 158 LM_DBG("'%.*s' Not found in usrloc\n",
>> aor.len, ZSW(aor.s));
>> 159 ul.unlock_udomain(_d, &aor);
>> 160 return -1;
>> 161 }
>> 162
>>
>> This is the point where I would need expertise help, because it looks
>> like it uses the "short" AoR (without URI gruu parameters) according to the
>> logs and a -1 is returned. Afterwards there are the lines used to lookup
>> the pub and temp gruu but are not, as far as I understand, used because of
>> the return -1.
>>
>> What is my mistake in the above assumption?
>>
>> Thanks a lot for the amazing fast reply.
>>
>> Samuel.
>>
>>
>>
>> On 26 August 2014 18:22, Daniel-Constantin Mierla <miconda at gmail.com>
>> wrote:
>>
>>> Hello,
>>>
>>> can you send a trace that includes the registration as well as the call?
>>>
>>> The pub-gruu is using the AoR, iirc.
>>>
>>> Also, the line you refer to is not matching anymore with latest 4.1.x --
>>> paste the code around it to locate it properly.
>>>
>>> Cheers,
>>> Daniel
>>>
>>>
>>> On 26/08/14 18:05, samuel wrote:
>>>
>>> Hi all,
>>>
>>> I'm having some issues treating requests within dialogs with gruu
>>> enabled with kamailio 4.1.2.
>>>
>>> I've got the "standard" configuration of WITHIN route with the adition
>>> of the next lines:
>>>
>>> if(is_gruu()){
>>> route(LOCATION);
>>> };
>>>
>>> before the the RELAY route call in the loose_route section.
>>>
>>> The "problem" is that the ACK with a pub-gruu on the Req-URI is not
>>> properly lookup. In the logs I can see the following statements:
>>> 2(4232) DEBUG: registrar [lookup.c:123]: lookup(): looking up pub gruu
>>> [urn:uuid:d63b1c4f-d7dc-4f4e-87f1-948123266dc0]
>>> 2(4232) DEBUG: registrar [lookup.c:158]: lookup(): 'sam at A.B.C.D' Not
>>> found in usrloc
>>>
>>> Where A.B.C.D is the local IP of the UA.
>>>
>>> Looking at the code, this last line looks like is looking for the
>>> "standard" URI (username at domain) instead of using the pub gruu. Am I
>>> right with this assumption or am I missing something from the code?
>>> As far as I could look, it looks like there's an exit -1 statement in
>>> the line 158 of lookup.c which disables the following gruu treatment.
>>>
>>> Since the username with IP is not registered, this ACK is lost and the
>>> sesion is not stablished (lost ACK).
>>>
>>> Can anyone provide some hints why is this failing?
>>>
>>> Thanks a lot in advance!
>>> Samuel.
>>>
>>>
>>>
>>> _______________________________________________
>>> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing listsr-users at lists.sip-router.orghttp://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>>>
>>>
>>> --
>>> Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
>>> Next Kamailio Advanced Trainings 2014 - http://www.asipto.com
>>> Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
>>>
>>>
>>> _______________________________________________
>>> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
>>> sr-users at lists.sip-router.org
>>> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>>>
>>>
>>
>
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing listsr-users at lists.sip-router.orghttp://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>
>
> --
> Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
> Next Kamailio Advanced Trainings 2014 - http://www.asipto.com
> Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
>
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
> sr-users at lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20140902/a47b4341/attachment.html>
More information about the sr-users
mailing list