- sr-users - kamailio.org

Info: sipexer v1.2.0 released
by Daniel-Constantin Mierla 11 Feb '25

11 Feb '25

Hello, a short note to announce that that sipexer v1.2.0 was released: - https://github.com/miconda/sipexer/releases/tag/v1.2.0 It is a modern and flexible SIP testing cli tool that can be used to send SIP signalling messages over UDP/TCP/TLS/WebSocket. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com Kamailio Advanced Training, March 24-28, 2025 -- asipto.com Kamailio World Conference, May 12-13, 2025, Berlin -- kamailioworld.com

1 0

DNS Unlinked Warnings
by John Larsen 10 Feb '25

10 Feb '25

Hi Everyone, We recently attempted an upgrade from kamailio 5.5.7 to 5.7.6. We immediately started to receive a log of warning logs similar to `dns_hash_put_shm_unsafe(): unlinked item 0x7f08799dcd90` The code shows that this warning log was added to the DNS cache in september 2023 - but its unclear what it means to have an unlinked item exactly? Is anyone able to shed any light on how DNS entries might become unlinked, and what the implications of it are? thanks John Larsen John Larsen Tech Lead EJohn.Larsen(a)symbio.global W symbio.global This communication is intended only for the person to whom it is addressed and may contain confidential material. If you received this communication in error, please inform the sender immediately and delete all copies. Please think of the environment before printing this email.

2 1

switch between "sips:" and "sip:" schema
by James Lipski 10 Feb '25

10 Feb '25

Hello, I have a kamailio installation where it's basically doing transport protocol translation. To go into more detail, I have a UA communicating over TLS towards kamailio/ and kamailio forwards the request to a freeswitch installation via UDP, so call-flow is as follows: UA <-- (TLS) --> Kamailio <-- (UDP) --> Freeswitch For the most part, this does work/ it's only when I use a UA that uses "sips:" as the schema instead of "sip:" where issues occur -- in my current topology, freeswitch has some weird behavior where it doesn't disconnect the call properly. At least based on my testing, I believe the issue is relating to the schema -- as the UA in question does have the ability to switch between schemas/ when using "sip:" instead of "sips:" over TLS, no issues with the same call-flow. In the event that I do come across a UA that doesn't have that ability, I was wondering if kamailio has the ability to switch schemas when it sends traffic over the UDP connection. I'm able to replace most aspects of the packet with "sip:" however the only header which contains the "sips:" schema is the contact header. With this alone, it propagates to the B-leg side (UDP side) of the call; and will cause freeswitch to not disconnect calls properly. I'm using the topos module to obfuscate my topology, not sure if that matters. -------- For example, on the A-leg side of the call, kamailio receives this: INVITE sips:12125551000@dev-sip.server.com:5061 SIP/2.0 Via: SIP/2.0/TLS 172.16.1.194:5061;branch=z9hG4bK236557042;rport;alias From: <sips:100000000@dev-sip.server.com>;tag=444048970 To: <sips:12125551000@dev-sip.server.com> Call-ID: 1326419916-5061-4(a)BHC.BG.B.BJE CSeq: 21 INVITE Contact: <sips:100000000@172.16.1.194:5061;transport=tls> Proxy-Authorization: Digest username="100000000", realm="dev-sip.server.com", nonce="Z6Ig0WeiH6Vdxtvej0I3XGCTUAEfh+0/", [uri="sips:12125551000@dev-sip.server.com](mailto:uri=)", response="43c1487d077db36a70fffdd1e25b3b52", algorithm=MD5 Max-Forwards: 70 User-Agent: Grandstream HT812 1.0.57.1 Supported: replaces, path, timer, eventlist Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Type: application/sdp Accept: application/sdp, application/dtmf-relay Content-Length: 1034 v=0 o=100000000 8000 8000 IN IP4 172.16.1.194 s=SIP Call c=IN IP4 172.16.1.194 t=0 0 m=audio 5004 RTP/SAVP 0 101 a=sendrecv a=rtpmap:0 PCMU/8000 a=ptime:20 a=rtpmap:101 telephone-event/8000 a=crypto:1 AEAD_AES_256_GCM inline:UlX3PAjNBP/aQwtwyf5ymsfJciB+VdFxMElO5zLVooUqmcEzZ8YyQQk9stI=|2^32 ------------------------------------------------------ On the B-leg side of the call, kamailio forwards the following to freeswitch: INVITE sip:12125551000@fs.server.com SIP/2.0 Via: SIP/2.0/UDP 10.64.52.113:5080;branch=z9hG4bK1552.89ec9666294034a5611f908f2a6a427e.0;i=f From: "Test" <sip:16461234567@dev-sip.server.com>;tag=444048970 To: <sip:12125551000@fs.server.com> Call-ID: 1326419916-5061-4(a)BHC.BG.B.BJE CSeq: 21 INVITE Max-Forwards: 15 Supported: replaces, path, timer, eventlist Allow: INVITE, ACK, OPTIONS, CANCEL, BYE, SUBSCRIBE, NOTIFY, INFO, REFER, UPDATE Content-Type: application/sdp Accept: application/sdp, application/dtmf-relay Content-Length: 316 Contact: <sips:btpsh-67a21ca2-2dbe86-1@10.64.52.113:5080> v=0 o=100000000 8000 8000 IN IP4 10.64.52.113 s=SIP Call c=IN IP4 10.64.52.113 t=0 0 m=audio 13060 RTP/AVP 0 101 a=rtpmap:0 PCMU/8000 a=rtpmap:101 telephone-event/8000 a=sendrecv a=ptime:20 -------------------------------------------------- As with the above, I was wondering if there is a way within kamailio to switch schemas on the b-leg side of this call example; specifically is there a way to update the contact header so it doesn't use the "sips:" schema when it traverses the UDP connection. Thank you for you time.

4 5

Auth_ephemeral module setup
by Anoop L 04 Feb '25

04 Feb '25

Hi, I'm trying to understand and setup Auth_ephemeral for authentication. My goal is to authenticate users dynamically using an web service for that I bumped into auth_ephemeral module. In the module how is a request made to the web service? Are username and password fetched from Pseudo-Variables variable? Is this module to be worked with xhttp module for http requests? Any help is much appreciated.

1 0

Auth_ephemeral Module Setup
by anoop96115＠outlook.com 04 Feb '25

04 Feb '25

Hi, I'm trying to understand and setup Auth_ephemeral for authentication. My goal is to authenticate users dynamically using an web service for that I bumped into auth_ephemeral module. In the module how is a request made to the web service? Are username and password fetched from Pseudo-Variables variable? Is this module to be worked with xhttp module for http requests? Any help is much appreciated.

1 0

Testing Kamailio IMS Modules with IPSec as P-CSCF
by Saeed Mogharian 03 Feb '25

03 Feb '25

Hi, I’m looking for suggestions on how to test Kamailio IMS modules, specifically for the IMS registration process over IPSec when Kamailio is configured as the P-CSCF node. Are there any scripts or IMS clients available that can help with this? Thanks in advance!

2 1

Kamailio 6.0 presentation at FOSDEM 2025
by Henning Westerholt 31 Jan '25

31 Jan '25

Hello, just a quick announcement, that tomorrow afternoon will be a presentation about Kamailio 6.0 at the FOSDEM conference in Brussels, Belgium [1]. Usually all the talks in this room are streamed online as well. I am happy also to meet during the conference to chat about Kamailio or other topics, just send me a (direct) message. Best regards, Henning [1] https://fosdem.org/2025/schedule/event/fosdem-2025-5978-kamailio-6-0-develo… -- Henning Westerholt - https://skalatan.de/blog/ Kamailio services - https://gilawa.com<https://gilawa.com/>

1 0

Strange memory stats
by Barry Flanagan 30 Jan '25

30 Jan '25

Hi, I have a 5.8.4 runing on Bookworm that after a while reports very strange memory stats like this: shmem.fragments: '729' shmem.free_size: '18446744009887078560' shmem.max_used_size: '67043934736' shmem.real_used_size: '67043698528' shmem.total_size: '3221225472' shmem.used_size: '67034971568' The settings are: SHM_MEMORY=3072 PKG_MEMORY=256 The shmem.free_size and max_used/max_size seem quite wrong. The output of "mod.stats all shm" shows the below: - 'Module: jsonrpcs' - Total: 0 - 'Module: core' - Total: 7891296 - 'Module: rr' - Total: 0 - 'Module: dialog' - Total: 146144 - 'Module: permissions' - Total: 130672 - 'Module: auth' - Total: 0 - 'Module: topoh' - Total: 0 - 'Module: sl' - Total: 14048 - 'Module: siptrace' - Total: 79184 - 'Module: sqlops' - Total: 0 - 'Module: acc' - Total: 0 - 'Module: xlog' - Total: 0 - 'Module: dmq' - Total: 1520 - 'Module: avpops' - Total: 0 - 'Module: rtpengine' - Total: 69328 - 'Module: db_mysql' - Total: 0 - 'Module: geoip2' - Total: 32 - 'Module: pv' - Total: 0 - 'Module: registrar' - Total: 0 - 'Module: textops' - Total: 0 - 'Module: htable' - Total: 6375584 - 'Module: tsilo' - Total: 74048 - 'Module: usrloc' - Total: 554240 - 'Module: tm' - Total: 6812224 - 'Module: websocket' - Total: 2551536 - 'Module: pike' - Total: 6880 - 'Module: tls' - Total: 5462960 - 'Module: tmx' - Total: 21968 - 'Module: kex' - Total: 3536 - 'Module: cfgutils' - Total: 128 - 'Module: sipdump' - Total: 32 - 'Module: domain' - Total: 5920 - 'Module: dispatcher' - Total: 20144 Thanks! -Barry

2 8

Kamailio v6.0.0 Released - new major version is out
by Daniel-Constantin Mierla 29 Jan '25

29 Jan '25

Kamailio v6.0.0 is out – it comes with 5 new modules and important updates to the core components and existing modules. The increment of the first version number reflects several important updates, among them: multi-threaded option for UDP receiving, option for specifying a range of ports to listen to, support for expressions with variables for most of the module functions parameters, internal libraries are directly inked to the core, support to use cmake to compile and install Kamailio. You can read a bit more detailed release notes at: * https://www.kamailio.org/w/kamailio-v6.0-0-release-notes/ Many thanks to all developers and community members that made possible this release. v6.0.0 brings more flexibility and optimizations across many existing components and modules, new parameters, functions, variables and transformations. Enjoy Kamailio v6.0.0! Thank you for flying Kamailio! Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com Kamailio World Conference, May 12-13, 2025, Berlin -- kamailioworld.com

2 2

Releasing v5.8.5
by Daniel-Constantin Mierla 29 Jan '25

29 Jan '25

Hello, I am considering to release Kamailio v5.8.5 (out of branch 5.8) on Thursday, Jan 23, 2025. If anyone is aware of issues not yet on the bug tracker, report them there asap in order to have a better chance to be fixed. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com

1 2

MS Teams BYE routing
by Barry Flanagan 28 Jan '25

28 Jan '25

Hi all, I have a Teams gw working apart from one issue., In a call originated in Teams,. if the called party hangs up, the BYE is sent to api-du-b-euwe.pstnhub.microsoft.com:443 which does not work. This is taken from the Contact header of the ACK I presume. Here is the ACK to the 200 OK: 28/01/2025 17:08:26.560 +00:00: 52.114.75.24:53728 -> 100.100.198.241:5061 ACK sip:9000@teams.domain.com:5061;transport=tls;tps=atpsh-67990f01-8ed8-1 SIP/2.0 FROM: "Teams User" <sip:+35314375362@sip.pstnhub.microsoft.com:5061;user=phone>;tag=c29cb7a79ddf40f9ba909f3f1034a770 TO: <sip:9000@teams.domain.com:5061>;user=phone;tag=as6e287620 CSEQ: 1 ACK CALL-ID: e49f7df631ee576286ecf159e81952fe MAX-FORWARDS: 70 VIA: SIP/2.0/TLS 52.114.75.24:5061;branch=z9hG4bK46d28f33 CONTACT: <sip:api-du-b-euwe.pstnhub.microsoft.com:443;x-i=0781e8a1-91e6-4f6a-bc4a-73d6890827d4;x-c=e49f7df631ee576286ecf159e81952fe/d/8/703117d8337e4975a8216a54f59035c2> CONTENT-LENGTH: 0 USER-AGENT: Microsoft.PSTNHub.SIPProxy v.2025.1.20.11 i.EUWE.7 ALLOW: INVITE,ACK,OPTIONS,CANCEL,BYE,NOTIFY ..and here is the BYE. 28/01/2025 17:08:36.119 +00:00: 100.100.198.241:5060 -> 52.123.134.13:443 BYE sip:api-du-b-euwe.pstnhub.microsoft.com:443;x-i=0781e8a1-91e6-4f6a-bc4a-73d6890827d4;x-c=e49f7df631ee576286ecf159e81952fe/d/8/703117d8337e4975a8216a54f59035c2 SIP/2.0 Record-Route: <sip:teams.domain.com:5061;transport=tls;ftag=as6e287620;lr;r2=on> Record-Route: <sip:100.100.198.241:5060;ftag=as6e287620;lr;r2=on> Via: SIP/2.0/UDP 100.100.198.241;branch=z9hG4bKc89c.0c1cf04a771c30293432e5f1f59331c8.0 Via: SIP/2.0/UDP 102.102.139.182:5060;rport=5060;branch=z9hG4bK3c42c163 Max-Forwards: 69 From: <sip:9000@teams.domain.com:5061>;user=phone;tag=as6e287620 To: "Teams User"<sip:+35314375362@sip.pstnhub.microsoft.com:5061;user=phone>;tag=c29cb7a79ddf40f9ba909f3f1034a770 Call-ID: e49f7df631ee576286ecf159e81952fe CSeq: 102 BYE Content-Length: 0 How can I get around this? Using a pretty standard WITHINDLG route. I am using topos but not manipulating the Contact header. Thanks! -Barry

4 8

RTPengine module - pause/resume recording
by Who AmI 28 Jan '25

28 Jan '25

Hi list, I have been looking at the recording daemon in RTPengine with Kamailio controlling this via rtpengine_manage() flags: "record-call=on" in the initial Invite, or 200OK based on direction. I tested and this works brilliantly for the full calls, but when looking at pause/resume the module documentation is bare and I am a bit confused. It looks like the ngcp docs and Opensips module docs (for RTPengine) describe the pause/resume as just another start/stop flag. Use case is simple, all calls are recorded, but we want the option to pause, then resume once any private or personal information exchange has completed. My question is also simple - Is this the recommended way to pause and then resume recordings or have I missed something? The Kamailio documentation doesn't explicitly say this ( https://www.kamailio.org/docs/modules/stable/modules/rtpengine.html#rtpengi…) so just looking for advice if this is the best route to go. Thanks, John.

2 2

Re: [sr-dev] Re: Branching 6.0 series
by Daniel-Constantin Mierla 28 Jan '25

28 Jan '25

Hello, On 20.01.25 08:25, Daniel-Constantin Mierla via sr-dev wrote: > Hello, > > I think we should branch it on Wednesday, Jan 22, 2025, no matter on > what stage we are with cmake support. The old-Makefiles should be kept > anyhow (as they are or in a special folder with an easy way to recover > them), because there are many bits and pieces that can be discovered > later when more users. > > Because there were no many bugs reported to the C code specific to 6.0, > we can aim releasing on January 29, 2025, with old-Makefile still to be > used for tasks that are not yet covered by cmake. shall we still aim for releasing v6.0.0 on this upcoming Wednesday, January 29, 2025? We the old-Makefiles kept, it is doable, the wiki pages needs to be created for this release series, which I can do tomorrow, as I waited to see if how it should be the recommended method for compilation/installation. If the feeling is that we should wait, then probably the target date has to be shifted somewhere in the week starting on February 10, 2025, because Fosdem follows this weekend and then some countries have a week holiday break. If no other opinions and change of decision, the release will be done on Wednesday, with content using the old-Makefiles (to reuse existing tutorials) and reference to cmake variant when it is the case. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com Kamailio World Conference, May 12-13, 2025, Berlin -- kamailioworld.com

2 1

defining configuration variables
by Sergio Charrua 27 Jan '25

27 Jan '25

Hello, just wondering if there is any way to define 2 constants as : # sets quarantine period the URL is set, in secs, after number of HTTP_RETRIES retries #!define SOME_VAR_1 60 # sets the configuration values for HTable module parameter #!define SOME_VAR_2 "some_configration_value" + SOME_VAR_1 I couldn't find anything in docs, and made several different ways without success. Any suggestions? Atenciosamente / Kind Regards / Cordialement / Un saludo, *Sérgio Charrua*

2 2

UDP worker lock investigation
by Володимир Іванець 26 Jan '25

26 Jan '25

Hello everyone! I was wondering if it is possible to get information about a packet that is being possessed by the worker without increasing log level? I have an installation with Kamailio that is configured with two interfaces. Once in a while Kamailio stops accepting requests on one of the interfaces. It works fine for requests from other one. I think that worker processes might be locked while processing some requests. Enabling debug log is not an option because the traffic is quite high and the log file will not just be enormous, some lines will be missing. It looks like syslog will skip some of the log entries. So is it possible to check what packet worker process is currently processing? Or do I need to add some log entries with lower log level to trace requests manually? Thank you!

3 4

How to access registration contact user agent string in branch route after location lookup?
by Benoit Panizzon 24 Jan '25

24 Jan '25

Hi List I have some CPE which require special handling like removing 100rel and update from the supported methods because they misbehave. Previously, I was using reg_fetch_contacts and then appending the branches manually and stacking the UA in an AVP. With the branch index this allowed me to get the AVP containing the UA to which the call was branched in the branch route and perform filtering of methods based on UA, but it was complicated. Now I use the intended function lookup_to_dset() but now I miss how to access the register contact UA from within a branch route. Is there an intended way to access that information? Mit freundlichen Grüssen -Benoît Panizzon- -- I m p r o W a r e A G - Leiter Commerce Kunden ______________________________________________________ Zurlindenstrasse 29 Tel +41 61 826 93 00 CH-4133 Pratteln Fax +41 61 826 93 01 Schweiz Web http://www.imp.ch ______________________________________________________

3 7

Use local_identity or custom data to set callerID
by matu lopi 23 Jan '25

23 Jan '25

Hi team, Is it possible to use either local_identity or custom data properties to define the CallerID during the RTCSession lifecycle? Thank you.

2 1

Issue with ACK on INVITE
by Barry Flanagan 23 Jan '25

23 Jan '25

Hi folks, Trying to hook up a new endpoint but am having an issue. Kamailio is in front of an Asterisk box. They send an INVITE, we send 100, then 200 OK. However, when they send their ACK, the RURI is not set to the Contact of the 200, instead it is<number>@<proxy ip>. This causes the ACK to get routed to the proxy itself and the call fails. 82.160.190.100:5060 = Kamailio IP Port 198.201.240.242:5080 = Asterisk IP/Port 182.33.174.10:5060 = Provider endpoint The far endpoint say they cannot fix the RURI - should I be able to handle this ACK below? My understanding is the ACK's RURI should be the Contact of the 200 OK. 200 OK sent from us to the Provider (Contact shows correct URI) ========================================================= 2025/01/15 17:14:49.470634 82.160.190.100:5060 -> 182.33.174.10:5060 SIP/2.0 200 OK Via: SIP/2.0/UDP 182.33.174.10;branch=z9hG4bKefc7.ad07f2d3.0 Via: SIP/2.0/UDP 182.33.174.39:5060;received=182.33.174.39;branch=z9hG4bK5bbf9591;rport=5060 Record-Route: <sip:1800715080@82.160.190.100:5061;r2=on;lr=on;ftag=as0b42eef3;did=f2e.ade2;nat=yes> Record-Route: <sip:1800715080@82.160.190.100;r2=on;lr=on;ftag=as0b42eef3;did=f2e.ade2;nat=yes> Record-Route: <sip:182.33.174.10;lr;ftag=as0b42eef3> From: "0737965510" <sip:0737966610@182.33.174.39>;tag=as0b42eef3 To: <sip:1800715080@82.160.190.100>;tag=as4133584e Call-ID: 480dc17b3f88516f364778dc4b2528da@182.33.174.39:5060 CSeq: 102 INVITE Server: Asterisk PBX 18.9.0 Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO, PUBLISH, MESSAGE Supported: replaces *Contact: <sip:1800715080@198.201.240.242:5080>* Content-Type: application/sdp Content-Length: 314 , ACK Reply from Provider to Us (RURI points to Kamailio ip:port now =========================================================== 2025/01/15 17:14:49.478119 182.33.174.10:5060 -> 82.160.190.100:5060 *ACK sip:1800715080@82.160.190.100:5060 *SIP/2.0 Via: SIP/2.0/UDP 182.33.174.10;branch=z9hG4bKefc7.ad07f2d3.2 Via: SIP/2.0/UDP 182.33.174.39:5060;received=182.33.174.39;branch=z9hG4bK55656469;rport=5060 Route: <sip:1800715080@82.160.190.100;r2=on;lr=on;ftag=as0b42eef3;did=f2e.ade2;nat=yes>,sip:1800715080@82.160.190.100:5061;r2=on;lr=on;ftag=as0b42eef3;did=f2e.ade2;nat=yes> Max-Forwards: 69 From: "0737965510" <sip:0737966610@182.33.174.39>;tag=as0b42eef3 To: <sip:1800715080@82.160.190.100>;tag=as4133584e Contact: <sip:0737965510@182.33.174.39:5060> Call-ID: 480dc17b3f88516f364778dc4b2528da@182.33.174.39:5060 CSeq: 102 ACK Thanks -Barry

5 5

Kamailio v5.8.5 Released
by Daniel-Constantin Mierla 23 Jan '25

23 Jan '25

Hello, Kamailio SIP Server v5.8.5 stable release is out. This is a maintenance release of the latest stable branch, 5.8, that includes fixes since the release of v5.8.4. There is no change to database schema or configuration language structure that you have to do on previous installations of v5.8.x. Deployments running previous v5.8.x versions are strongly recommended to be upgraded to v5.8.5. For more details about version 5.8.5 (including links and guidelines to download the tarball or from GIT repository), visit: * https://www.kamailio.org/w/2025/01/kamailio-v5-8-5-released/ RPM, Debian/Ubuntu packages will be available soon as well. Many thanks to all contributing and using Kamailio! Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com Kamailio World Conference, May 12-13, 2025, Berlin -- kamailioworld.com

1 0

Getting headers from xhttp/websocket
by Barry Flanagan 23 Jan '25

23 Jan '25

Hi, I need to be able to pull in a header provided by a webrtc endpoint so I can use it in my request_route. I understand that I will need to use shm/htable to make it visible but I am unable to get the xhttp header using $header(x-p-push), or any header by name. In my event_route[xhttp:request] I have the following: if ($hdr(Upgrade) =~ "websocket" && $hdr(Connection) =~ "Upgrade" && $rm =~ "GET") { .... xlog("L_INFO", "BF x-p-push header is $hdr(x-p-push) Contact is $hdr(Contact)\n"); } The log shows both x-p-push and Contact header as null. sipdump shows the headers present in the request. Am I missing something? Running Kamailio 5.8.4 Thanks -Barry

3 3

Development is open for future release of v6.1.x series
by Daniel-Constantin Mierla 22 Jan '25

22 Jan '25

Hello, the branch 6.0 was created, therefore the master branch is open for adding new features, to be part of future release series v6.1.x. Any bug fix committed to master that applies to 6.0.x or older stable branches should be backported as usual with "git cherry-pick -x ..." to appropriate branches like 6.0 or 5.8. Expect that v6.0.0 will be released in a few weeks from now. Based on the workflow used during the past years, the next future release v6.1.0 should be out after another 8-10 months of development, plus 1-2 months of testing, so sometime during the last part of 2025 or the beginning of 2026. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com Kamailio World Conference, May 12-13, 2025, Berlin -- kamailioworld.com

1 0

Branch 6.0 has been created
by Daniel-Constantin Mierla 22 Jan '25

22 Jan '25

Hello, the branch 6.0 has been created, to be used for releasing v6.0.x series. To check out this branch, the following commands can be used: git clone https://github.com/kamailio/kamailio kamailio-6.0 cd kamailio-6.0 git checkout -b 6.0 origin/6.0 Pushing commits in this branch: git push origin 6.0:6.0 Note that 6.0 is an official stable branch, so only bug fixes, missing kemi exports (discuss on sr-dev if not sure) or improvements to documentation or helper tools can be pushed to this branch. As usual, if there is a bug fixed, commit and push first to master branch and then cherry pick to 6.0 branch: git cherry-pick -x COMMITID In few weeks, the first release from branch 6.0 will be out, respectively Kamailio v6.0.0. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com

1 0

Not able to get the $TV(un) inside xhttp module
by kaseenadh111＠gmail.com 22 Jan '25

22 Jan '25

Hi, I am running Kamailio 5.7.* RHEL I need to log the latency of each XHTTP request (milli or microseconds precision) so I tried using $TV(un) which always gives a 0 value. Can we access $TV(un) under XHTTP routes? event_route[xhttp:request] { $var(tv_http_req_start_time) = $TV(un); xlog("L_INFO", "$$var(tv_http_req_start_time)= $var(tv_http_req_start_time) and actual $$TV(un)= $TV(un)"); .... } Output: Jan 22 16:18:20 VM-123************ kamailio[346535]: INFO: <script>: $var(tv_http_req_start_time)= 0 and actual $TV(un)= 0 $TS - I have tried this, it gives the Unix in seconds precision but I want milliseconds precision. Thanks Kasee

1 0

Re: [sr-dev] Re: Branching 6.0 series
by Daniel-Constantin Mierla 22 Jan '25

22 Jan '25

Hello, later today, in a few hours, I am going to create the branch 6.0. More details will be provided once it has been created. Cheers, Daniel On 20.01.25 08:25, Daniel-Constantin Mierla via sr-dev wrote: > Hello, > > I think we should branch it on Wednesday, Jan 22, 2025, no matter on > what stage we are with cmake support. The old-Makefiles should be kept > anyhow (as they are or in a special folder with an easy way to recover > them), because there are many bits and pieces that can be discovered > later when more users. > > Because there were no many bugs reported to the C code specific to 6.0, > we can aim releasing on January 29, 2025, with old-Makefile still to be > used for tasks that are not yet covered by cmake. > > Cheers, > Daniel > > On 15.01.25 11:16, Henning Westerholt via sr-dev wrote: >> Hello, >> >> yes, Xenofon is also waiting for a final reply from Sergey for the RPM packaging if now all is fine (was discussed as part of PR #4085). >> >> Cheers, >> >> Henning >> >>> -----Original Message----- >>> From: Victor Seva via sr-dev <sr-dev(a)lists.kamailio.org> >>> Sent: Mittwoch, 15. Januar 2025 10:59 >>> To: sr-dev(a)lists.kamailio.org >>> Cc: Victor Seva <linuxmaniac(a)torreviejawireless.org> >>> Subject: [sr-dev] Re: Branching 6.0 series >>> >>> Hi, >>> >>> please postpone. I didn't have time finalize the test of cmake builds on debian >>> packaging yet. >>> >>> On 15/1/25 7:18, Daniel-Constantin Mierla via sr-dev wrote: >>>> Hello, >>>> >>>> creating git branch 6.0 was planned for today and I wonder if it shall >>>> be done latter during the evening or it would be preferred to be >>>> postponed for two days or so, till end of the week/Friday, as I >>>> noticed yesterday some new work being done to cmake files for FreeBSD >>>> and maybe new commits are still coming. >>>> >>>> Of course, there is also the option to create the branch and then the >>>> new commits backported, but maybe postponing makes everything more >>>> convenient. >>>> >>>> In short, if anyone feels that postponing a bit creating the branch >>>> 6.0 simplifies their current work, let us know. If nobody ask for >>>> postponing, the branch will be created later towards the end of today. >>>> >>>> Cheers, >>>> Daniel >>> -- >>> ----------------------------------------------------------------- >>> | ⢀⣴⠾⠻⢶⣦⠀ Victor Seva | >>> | ⣾⠁⢠⠒⠀⣿⡁ linuxmaniac(a)torreviejawireless.org | >>> | ⢿⡄⠘⠷⠚⠋⠀PGP: 8F19 CADC D42A 42D4 5563 730C 51A0 9B18 CF5A >>> 5068 | >>> | ⠈⠳⣄⠀⠀⠀ Debian Developer | >>> ----------------------------------------------------------------- >> _______________________________________________ >> Kamailio - Development Mailing List -- sr-dev(a)lists.kamailio.org >> To unsubscribe send an email to sr-dev-leave(a)lists.kamailio.org >> Important: keep the mailing list in the recipients, do not reply only to the sender! > > -- > Daniel-Constantin Mierla (@ asipto.com) > twitter.com/miconda -- linkedin.com/in/miconda > Kamailio Consultancy, Training and Development Services -- asipto.com > > _______________________________________________ > Kamailio - Development Mailing List -- sr-dev(a)lists.kamailio.org > To unsubscribe send an email to sr-dev-leave(a)lists.kamailio.org > Important: keep the mailing list in the recipients, do not reply only to the sender! -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com

2 1

Multi-line Via: rport added incorrectly
by James Browne 22 Jan '25

22 Jan '25

Hi I see that when kamailio adds rport to the Via header field of a request that has two Via values on the same line (comma-separated, of course), it adds the rport (and received) to the wrong value. I have this test kamailio.cfg for demonstration. children=1 loadmodule "sl" loadmodule "textops" loadmodule "nathelper" request_route { force_rport(); sl_send_reply(200, "OK"); } I send this OPTIONS (truncated) and get this response (truncated). The rport should be on the first Via value, not the second. OPTIONS sip:server SIP/2.0 Via: SIP/2.0/UDP 3.9.4.2:5060;branch=z9hG4bK503983089,SIP/2.0/UDP 1.3.4.1:5060;branch=z9hG4bK53805983059 From: <sip:client>;tag=LeonhardEuler To: <sip:server> SIP/2.0 200 OK Via: SIP/2.0/UDP 3.9.4.2:5060;branch=z9hG4bK503983089,SIP/2.0/UDP 1.3.4.1:5060;branch=z9hG4bK53805983059;rport=53805;received=127.0.0.1 From: <sip:client>;tag=LeonhardEuler To: <sip:server>;tag=9dd61ff61e802d8e2bef5f14621ef3c2.49678e65 Server: kamailio (6.0.0-pre0 (x86_64/linux)) I've tested this with the latest commit (c994fb8) from kamailio. I can't think how this can be anything other than a bug. Should I log a bug report for this? James

2 4

Error when sending NOTIFYs to disconnected websocket clients
by Flavio Grossi 20 Jan '25

20 Jan '25

Hi, i have some SIP websocket clients (built using the SIP.js library) SUBSCRIBEd to other clients via kamailio v5.8.2. If one of the watchers suddenly disappears (e.g. the web page is reloaded), another connection is made and a new watcher appears in kamailio without the old one disappearing. For example: # kamctl rpc presence.watcher_list basic sip:1001@example.voismart.com { "jsonrpc": "2.0", "result": [ { "pres_uri": "sip:1001@example.com", "to_user": "1001", "watcher_user": "1000", "contact": "sip:1000@mfp0h2dhhmq2.invalid;transport=ws;alias=192.168.56.10~34166~5", "callid": "1o9b499kj6l65if0p30t", "user_agent": "sip.js/1.12.1", "expires": 1736422494, ... }, { "pres_uri": "sip:1001@example.com", "to_user": "1001", "watcher_user": "1000", "contact": "sip:1000@0j7qoabnv827.invalid;transport=ws;alias=192.168.56.10~34156~5", "callid": "8mjhrqifgm39up9t2qq3", "user_agent": "sip.js/1.12.1", "expires": 1736422492, ... } ], "id": 11753 } you can see the duplicated watcher in which the old connection is now invalid because the websocket client disappeared. In this situation, issuing a pres_refresh_watchers() call will fail to send any notify (even to the valid connection) because of this error: WARNING: <core> [core/msg_translator.c:3007]: via_builder(): TCP/TLS connection (id: 0) for WebSocket could not be found ERROR: tm [t_msgbuilder.c:1423]: assemble_via(): via building failed ERROR: tm [t_msgbuilder.c:1614]: build_uac_req(): error while assembling Via ERROR: tm [uac.c:552]: t_uac_prepare(): Error while building message ERROR: presence [notify.c:1734]: send_notify_request(): in function tm t_request_within() ERROR: presence [notify.c:1829]: notify(): sending Notify not successful ERROR: presence [notify.c:1585]: query_db_notify(): Could not send notify for [event]=dialog ERROR: presence [presence.c:739]: pres_refresh_watchers(): sending Notify requests in which the building of the notify message fails because it cannot find a valid connection when building the via header. Is this a wanted behaviour? shouldn't it just fail the sending of the single message? Is there anything i can do to fix this? Trying to reproduce the same problem with a regular tcp client, it seems that the other notifys are correctly sent and only a timeout error is sent for the single message: ERROR: <core> [core/tcp_main.c:4622]: handle_tcpconn_ev(): connect 192.168.56.10:21994 failed which i think is the correct behaviour. Related question: is it possible to remove active watchers when the connection goes down? Thanks for the help on this.

2 5

Performance issues with KEMI
by Daniel W. Graham 20 Jan '25

20 Jan '25

Testing out KEMI functionality and running into performance issues. If I exceed 150 calls per second the network receive queue grows and Kamailio is unable to keep up with requests and they begin dropping. KEMI script for testing is just doing a stateless reply to invites. Using python3s module. I've played with Kamailio child processes and memory allocations, but there is no impact. I've also attempted some buffer / memory tweaking at the OS level, again with no impact. Increasing CPU cores and even running the test on bare metal results in the same. Example of receive queue at 150 calls per second - Netid State Recv-Q Send-Q Local Address:Port udp UNCONN 337280 0 x.x.x.x:5060 Just wondering if anyone has experienced similar issues or has an example of the performance they are seeing before I continue down this path. Thanks, - dan

5 9

Need help with serial branching using: t_load_contacts(1)
by Benoît Panizzon 20 Jan '25

20 Jan '25

Hi Gang I guess, I don't completely understand who to properly perform serial branching. I did try to follow the examples from: https://www.kamailio.org/docs/modules/devel/modules/tm.html#tm.f.t_next_con… This is, stripped down, the relevant config used. modparam("tm", "contacts_avp", "tm_contacts") modparam("tm", "contact_flows_avp", "tm_contact_flows") route[LOCATION] { lookup_to_dset("location", "$var(lookupuri)"); if (!t_load_contacts(1)) { xlog("L_ERR", "$cfg(route): ######### load_contacts failed\n"); sl_send_reply("500", "Server Internal Error - Cannot load contacts"); exit; } xlog("L_INFO", "$cfg(route): Contacts loaded: $cnt($xavp(tm_contacts))\n"); => Confirms, there is more than one contact loaded. if (!t_next_contacts()) { send_reply("480", "Not registered"); } t_set_fr(5000,1500); # Set 5 second timeout for LAB testing to quickly try the next contact. t_on_branch("BR_TO_CUST"); t_on_branch_failure("BR_TO_CUST_FAILURE"); exit; } event_route[tm:branch-failure:BR_TO_CUST_FAILURE] { xlog("BRANCH FAILED $T_reply_code to $rm message\n"); t_on_branch("BR_TO_CUST"); t_on_branch_failure("BR_TO_CUST_FAILURE"); if (t_next_contact_flow()) { xlog("BRANCH FAILED, Try Next\n"); t_relay(); } else { xlog("L_INFO", "No more flows\n"); t_reply("408", "Branch Timeout"); exit; } } What happens is: INVITE is sent to the first contact, who is replying RINGING. After 5 seconds the timeout is reached and the branch-failure route engaged. Jan 17 15:01:15 dev-cpereg01 kamailio[3599432]: ERROR: <script>: BRANCH FAILED 408 to INVITE message Jan 17 15:01:15 dev-cpereg01 kamailio[3599432]: INFO: <script>: No more flows Jan 17 15:01:15 dev-cpereg01 kamailio[3599432]: CRITICAL: tm [tm.c:1554]: ki_t_reply(): w_t_reply entered in unsupported mode To my understanding, t_next_contact_flow() should have loaded the next ds, but this does not seem to happen. What am I missing? -- Mit freundlichen Grüssen -Benoît Panizzon- @ HomeOffice und normal erreichbar -- I m p r o W a r e A G - Leiter Commerce Kunden ______________________________________________________ Zurlindenstrasse 29 Tel +41 61 826 93 00 CH-4133 Pratteln Fax +41 61 826 93 01 Schweiz Web http://www.imp.ch ______________________________________________________

2 3

Kamailio IMS, used with IPv6?
by Valentin Christoph 16 Jan '25

16 Jan '25

Anybody experiences with Kamailio IMS modules (ims_registrar_scscf, ims_usrloc_scscf, ims_isc, ........) and the use of IPv6 addresses? Any pitfalls? DNS issues? Thanks, Christoph

2 2

Asynchttp failover
by Sergio Charrua 15 Jan '25

15 Jan '25

Hello all! just wondering if anyone knows a good way to failover with the HTTP Async module. The standard HTTP Client module (synchronous) does implement some nice functionalities for failover between URL addresses, but the HTTP Async doesn't have any failover functionality... I have implemented some functions (within kamailio.cfg) for failover upon HTTP requests, but despite it working fine, it would be better to have that functionality on the HTTP Async module, just like HTTP Client has... Any technical reason for HTTP Async module not having failover functionalities? Atenciosamente / Kind Regards / Cordialement / Un saludo, *Sérgio Charrua*

2 2

Module snmpstats reporting inconsistent call success metrics on intense peaks of failed calls
by Igor Potjevlesch 15 Jan '25

15 Jan '25

Hello! We are plotting the call success rate on each of our proxies. Using the Kamailio snmpstats module, we try to obtain three counters to plot them on a graph: Total Calls Active Calls Error Calls Most of the time, everything works well, and we obtain sensible values: snmpwalk -c mycommunity -v 2c W.X.Y.Z 1.3.6.1.4.1.34352.3.1.3.1.3.2 -O n .1.3.6.1.4.1.34352.3.1.3.1.3.2.1.0 = Gauge32: 309 -> Total Calls .1.3.6.1.4.1.34352.3.1.3.1.3.2.2.0 = Gauge32: 95 -> Active Calls .1.3.6.1.4.1.34352.3.1.3.1.3.2.3.0 = Gauge32: 214 -> Error calls However, when our proxies are facing intense peaks of failed calls in a short time span, sometimes, the "Error Calls" counter returns a value higher than "Total Calls". As a result, "Active Calls" suddenly reaches a value close to the maximum of a Gauge32 (because T o t a l C a l l s − E r r o r C a l l s < 0 ). snmpwalk -c mycommunity -v 2c W.X.Y.Z 1.3.6.1.4.1.34352.3.1.3.1.3.2 -O n .1.3.6.1.4.1.34352.3.1.3.1.3.2.1.0 = Gauge32: 153 -> Total Calls .1.3.6.1.4.1.34352.3.1.3.1.3.2.2.0 = Gauge32: 4294967226 -> Active Calls .1.3.6.1.4.1.34352.3.1.3.1.3.2.3.0 = Gauge32: 223 -> Error calls As a result, our graphs make no more sense. We are currently running kamailio 5.6.4. Any idea if it's a bug or a configuration issue? Regards, Igor. -- Cet e-mail a été vérifié par le logiciel antivirus d'Avast. www.avast.com

3 7

Branching 6.0 series
by Daniel-Constantin Mierla 15 Jan '25

15 Jan '25

Hello, I am trying to figure out what would be the best date to branch 6.0 in the git repo. To allow a bit of time after returning from the winter holidays to tune more the cmake and packaging, I would think that January 15, 2025 could be a good choice. If nobody comes with other opinions/suggestions, then we will go for it. Cheers, Daniel -- Daniel-Constantin Mierla (@ asipto.com) twitter.com/miconda -- linkedin.com/in/miconda Kamailio Consultancy, Training and Development Services -- asipto.com

1 1

kamcmd tls.reload issue
by Yuriy Nasida 14 Jan '25

14 Jan '25

Hello, I am using letsencrypt cert and key and do not want to restart kamailio every 3 months to load new ones. I know that there is: kamcmd tls.reload method but it has an error for me. error: 500 - Error while fixing TLS configuration (consult server log) I am checking the logs and see: kamailio[3865480]: INFO: tls [tls_domain.c:345]: ksr_tls_fill_missing(): TLSs<default>: tls_method=3 kamailio[3865480]: INFO: tls [tls_domain.c:357]: ksr_tls_fill_missing(): TLSs<default>: certificate='/etc/kamailio/certs/my_cert.crt' kamailio[3865480]: INFO: tls [tls_domain.c:364]: ksr_tls_fill_missing(): TLSs<default>: ca_list='(null)' kamailio[3865480]: INFO: tls [tls_domain.c:371]: ksr_tls_fill_missing(): TLSs<default>: ca_path='(null)' kamailio[3865480]: INFO: tls [tls_domain.c:378]: ksr_tls_fill_missing(): TLSs<default>: crl='(null)' kamailio[3865480]: INFO: tls [tls_domain.c:382]: ksr_tls_fill_missing(): TLSs<default>: require_certificate=0 kamailio[3865480]: INFO: tls [tls_domain.c:390]: ksr_tls_fill_missing(): TLSs<default>: cipher_list='(null)' kamailio[3865480]: INFO: tls [tls_domain.c:397]: ksr_tls_fill_missing(): TLSs<default>: private_key='/etc/kamailio/certs/private.key' kamailio[3865480]: INFO: tls [tls_domain.c:401]: ksr_tls_fill_missing(): TLSs<default>: verify_certificate=0 kamailio[3865480]: INFO: tls [tls_domain.c:406]: ksr_tls_fill_missing(): TLSs<default>: verify_depth=9 kamailio[3865480]: INFO: tls [tls_domain.c:410]: ksr_tls_fill_missing(): TLSs<default>: verify_client=0 kamailio[3865480]: NOTICE: tls [tls_domain.c:1168]: ksr_tls_fix_domain(): registered server_name callback handler for socket [:0], server_name='<default>' ... kamailio[3865480]: ERROR: tls [tls_domain.c:590]: load_cert(): TLSs<default>: Unable to load certificate file '/etc/kamailio/certs/my_cert.crt' kamailio[3865480]: ERROR: tls [tls_util.h:49]: tls_err_ret(): load_cert:error:03000072:digital envelope routines::decode error (sni: unknown) kamailio[3865480]: ERROR: tls [tls_util.h:49]: tls_err_ret(): load_cert:error:0A00018F:SSL routines::ee key too small (sni: unknown) Any advice ? It's interesting that there are not any errors in case I restart kamailio. I can make TLS calls without problems. deb 12.5 version: kamailio 5.7.4 (x86_64/linux)

4 5

FOSDEM VoIP Dinner
by Torrey Searle 14 Jan '25

14 Jan '25

Hello all, As in previous years, I'm organising the VoIP dinner for FOSDEM (1 Feb @ 19:00) However unlike previous years (since the restaurant got upset about splitting the bill 30 ways). I will ask everyone to order their meal and pay in advance. (and I will pay the bill to the restaurant with my card) If you are wanting to attend the dinner please fill out the following form and follow the instructions for payment. https://form.jotform.com/250012543310033 Contact details are also in the form if you want to reach out to me with any further questions. Kind regards Torrey

5 4

ims_diameter_server not freeing shared memory
by marpett365＠gmail.com 09 Jan '25

09 Jan '25

Hi, I have run in to an issue when sending diameter requests using the ims_diameter_server module. Shared memory keeps increasing and are not being freed until restart of kamailio. This leads to all shared memory is being used and no more are able to be allocated. This only happens when the request is sent synchronously. When sent asynchronously shared memory is being freed as far as I can see. have anyone ran into this issue and found a remedy or can give some pointers how to proceed? Following functions have been used for sending requests. diameter_request([peer], appid, commandcode, message) diameter_request_async([peer], appid, commandcode, message) route[CHECK_LOCATION] { diameter_request("hss.ims.test.local", "16777217", "306", "[ " + "{ \"avpCode\":260, \"vendorId\":0, \"Flags\":64, \"list\":[ { \"avpCode\":266, \"vendorId\":0, \"Flags\":64, \"int32\":10415 }, { \"avpCode\":258, \"vendorId\":0, \"Flags\":64, \"int32\":16777217 } ] }, " + "{ \"avpCode\":277, \"vendorId\":0, \"Flags\":64, \"int32\":1 }, " + "{ \"avpCode\":283, \"vendorId\":0, \"Flags\":64, \"string\":\"ims.test.local\"}, " + "{ \"avpCode\":706, \"vendorId\":10415, \"Flags\":192, \"int32\":0 }, " + "{ \"avpCode\":700, \"vendorId\":10415, \"Flags\":192, \"list\": [ { \"avpCode\":601, \"vendorId\":10415, \"Flags\":192, \"string\":\"tel:" + $fU + "\" }]}, " + "{ \"avpCode\":707, \"vendorId\":10415, \"Flags\":192, \"int32\":0 }, " + "{ \"avpCode\":703, \"vendorId\":10415, \"Flags\":192, \"int32\":14 } ]"); xlog("L_INFO", "Sent Diameter request"); } Shared memory after restart kamcmd mod.stats cdp shm Module: cdp { save_peer_applications(779): 144 make_send_pipe(126): 320 dp_add_pid(92): 216 AAAAddRequestHandler(69): 40 cb_add(183): 8 cb_add(177): 32 cdp_sessions_init(178): 8 cdp_sessions_init(173): 8 cdp_sessions_init(165): 1024 cdp_sessions_init(157): 3072 cdp_sessions_init(149): 8 cdp_trans_init(67): 8 cdp_trans_init(60): 24 add_timer(150): 24 add_timer(145): 144 new_peer(77): 16 new_peer(75): 16 new_peer(72): 16 new_peer(69): 80 new_peer(66): 96 new_peer(60): 384 peer_manager_init(84): 8 peer_manager_init(83): 8 peer_manager_init(82): 8 peer_manager_init(79): 8 peer_manager_init(76): 16 worker_init(101): 16 worker_init(94): 128 worker_init(89): 32 worker_init(87): 32 worker_init(84): 8 worker_init(82): 48 timer_cdp_init(182): 8 timer_cdp_init(179): 16 diameter_peer_init_real(137): 8 diameter_peer_init_real(131): 16 diameter_peer_init_real(122): 16 diameter_peer_init_real(115): 8 diameter_peer_init_real(108): 8 diameter_peer_init_real(101): 8 diameter_peer_init_real(94): 8 new_routing_entry(88): 64 parse_dp_config(447): 16 parse_dp_config(348): 8 parse_dp_config(340): 16 parse_dp_config(333): 40 parse_dp_config(326): 144 quote_trim_dup(96): 456 new_dp_config(52): 184 Total: 7024 } Shared memory after 10 diameter requests Module: cdp { save_peer_applications(779): 144 AAACreateAVP(137): 5072 AAATranslateMessage(500): 1320 do_receive(600): 7392 make_send_pipe(126): 320 dp_add_pid(92): 216 AAAAddRequestHandler(69): 40 cb_add(183): 8 cb_add(177): 32 cdp_sessions_init(178): 8 cdp_sessions_init(173): 8 cdp_sessions_init(165): 1024 cdp_sessions_init(157): 3072 cdp_sessions_init(149): 8 cdp_trans_init(67): 8 cdp_trans_init(60): 24 add_timer(150): 24 add_timer(145): 144 new_peer(77): 16 new_peer(75): 16 new_peer(72): 16 new_peer(69): 80 new_peer(66): 96 new_peer(60): 384 peer_manager_init(84): 8 peer_manager_init(83): 8 peer_manager_init(82): 8 peer_manager_init(79): 8 peer_manager_init(76): 16 worker_init(101): 16 worker_init(94): 128 worker_init(89): 32 worker_init(87): 32 worker_init(84): 8 worker_init(82): 48 timer_cdp_init(182): 8 timer_cdp_init(179): 16 diameter_peer_init_real(137): 8 diameter_peer_init_real(131): 16 diameter_peer_init_real(122): 16 diameter_peer_init_real(115): 8 diameter_peer_init_real(108): 8 diameter_peer_init_real(101): 8 diameter_peer_init_real(94): 8 new_routing_entry(88): 64 parse_dp_config(447): 16 parse_dp_config(348): 8 parse_dp_config(340): 16 parse_dp_config(333): 40 parse_dp_config(326): 144 quote_trim_dup(96): 456 new_dp_config(52): 184 Total: 20808 } Shared memory after 170k requests do_receive(600): 131363752 AAATranslateMessage(500): 32326656 AAACreateAVP(137): 110067152 Total: 273764840 Br Martin

2 5

evapi not processing received messages
by Sergio Charrua 09 Jan '25

09 Jan '25

Hi all! I have implemented evapi module on Kamailio 5.8.4, calling a python script. Python script is listening to a socket. It receives messages (JSON) from Kamailio, makes an HTTP request somewhere, and sends the response back to Kamailio. The Python script is working fine: it connects to Kamailio, sends HTTP requests, and forwards a response back to Kamailio: I have confirmed using TCPDump that, in fact, the data is received and forwarded back. However, despite showing messages that the Python client script is correctly connected to Kamailio, the event_route[evapi:message-received] is never executed. Probably there is some detail that I am missing, but the docs and online info for evapi are very rare, and they all show very similar configuration. And as I wrote, I have confirmed using TCPDump that the data is received from Python client script to Kamailio node. What am I missing? This is my Kamailio cfg: [loads modules] [sets modparams] modparam("evapi","bind_addr","10.20.20.1:8888") debug=2 children=16 log_facility=LOG_LOCAL0 log_prefix="{$mt $hdr(CSeq) $ci} " disable_sctp = yes force_rport = yes rundir="/tmp" request_route { if ( is_method("ACK") ) { if ( t_check_trans() ) { t_relay(); } exit; } if(is_method("OPTIONS")){ sl_reply("200", "OK"); exit; } if(is_method("CANCEL")){ sl_reply("200","OK"); sl_reply("487","Request Terminated"); exit; } route(TOEVAPI); exit; } event_route[evapi:connection-new] { xlog("LINFO","new connection from [$evapi(srcaddr):$evapi(srcport)]\n"); if($evapi(srcaddr)!="10.20.0.1") { evapi_close(); exit; } } event_route[evapi:connection-closed] { xlog("LINFO","connection closed by $evapi(srcaddr):$evapi(srcport)\n"); } event_route[evapi:message-received] { xlog("LINFO","received [$evapi(msg)] from $evapi(srcaddr):$evapi(srcport)\n"); jansson_get("t-index", "$evapi(msg)", "$var(t-index)"); jansson_get("t-label", "$evapi(msg)", "$var(t-label)"); $var(evmsg) = $evapi(msg); xlog("L_INFO", "preparing to resume transaction for processing: $var(tindex) / $var(tlabel)\n"); t_continue("$var(tindex)", "$var(tlabel)", "EVAPIRESPONSE"); } route[EVAPIRESPONSE] { xlog("L_INFO", "resumed transaction for processing: $T(id_index) / $T(id_label)\n"); [do stuff] } route[TOEVAPI]{ xlog("L_INFO", "suspended transaction: $T(id_index) / $T(id_label)\n"); evapi_async_relay( [some JSON string] ); exit; } Thanks in advance! Atenciosamente / Kind Regards / Cordialement / Un saludo, *Sérgio Charrua*

3 6

Kamailio and Faxing
by Temilolu Fayomi 09 Jan '25

09 Jan '25

Hello, I am having some issues ensuring t38 faxes are being received reliably. I am currently using kamailio as an inbound sbc with freepbx as the endpoint. I understand that potentially stripping other codecs except for g711a/u makes inbound faxes reliable, but beyond that I am having trouble determining how to ensure reliable faxing. Is this something kamailio can handle (such as an error correction mode or pass through setting or module?) I would appreciate any insight as I have hit a wall when it comes to this issue. Thank you, Temi

3 3

Re: WS and rtpengine_manage
by Knserbrave 08 Jan '25

08 Jan '25

test

2 4

deb-archive.kamailio.org unreachable
by Floimair Florian 08 Jan '25

08 Jan '25

Hi! It seems that deb-archive.kamailio.org is currently down. It resolves to IP 76.9.245.162 on my system Can someone with access to that please have a look. Thank you FLORIAN FLOIMAIR Software Development - Symphony Cloud Services Commend International GmbH Saalachstrasse 51 5020 Salzburg, Austria [signature_2459749227] commend.com LG Salzburg / FN 178618z

2 2

Kamailio crash
by Mitesh Thakkar 07 Jan '25

07 Jan '25

Hi, I have recently migrated from Kamailio 5.2.5 to Kamailio 5.5.7. After a migration, the Kamailio crashed. Find the details below. Thanks in advance. Back trace: ========= Core was generated by `/usr/local/sbin/kamailio -DD -E -m 1941 -M 970 -f /etc/kamailio/kamailio.cfg --'. Program terminated with signal SIGSEGV, Segmentation fault. #0 0x00007f8e27a2be15 in OPENSSL_cleanse () from /lib/x86_64-linux-gnu/libcrypto.so.3 (gdb) bt #0 0x00007f8e27a2be15 in OPENSSL_cleanse () from /lib/x86_64-linux-gnu/libcrypto.so.3 #1 0x00007f8e278faa8a in CRYPTO_clear_free () from /lib/x86_64-linux-gnu/libcrypto.so.3 #2 0x00007f8e27838175 in BUF_MEM_free () from /lib/x86_64-linux-gnu/libcrypto.so.3 #3 0x00007f8e27bc0c98 in SSL_free () from /lib/x86_64-linux-gnu/libssl.so.3 #4 0x00007f8e27c8cd6e in tls_h_tcpconn_clean_f (c=0x7f8db0fa1e90) at tls_server.c:701 #5 0x000055f7f42cc6e1 in _tcpconn_free (c=0x7f8db0fa1e90) at core/tcp_main.c:1541 #6 0x000055f7f42e2f3c in tcpconn_put_destroy (tcpconn=0x7f8db0fa1e90) at core/tcp_main.c:3285 #7 0x000055f7f42e8759 in handle_tcp_child (tcp_c=0x7f8e27d88078, fd_i=-1) at core/tcp_main.c:3722 #8 0x000055f7f42f506f in handle_io (fm=0x7f8e27db1578, ev=1, idx=-1) at core/tcp_main.c:4539 #9 0x000055f7f42b6e49 in io_wait_loop_epoll (h=0x55f7f45ae240 <io_h>, t=5, repeat=0) at core/io_wait.h:1070 #10 0x000055f7f42f8bc3 in tcp_main_loop () at core/tcp_main.c:4824 #11 0x000055f7f4057fc2 in main_loop () at main.c:1856 #12 0x000055f7f4062670 in main (argc=10, argv=0x7ffc6d583528) at main.c:3052 OS: Ubuntu 22.04.5 LTS \n \l # kamailio --version version: kamailio 5.5.7 (x86_64/linux) 272fdd flags: USE_TCP, USE_TLS, USE_SCTP, TLS_HOOKS, USE_RAW_SOCKS, DISABLE_NAGLE, USE_MCAST, DNS_IP_HACK, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, TLSF_MALLOC, DBG_SR_MEMORY, USE_FUTEX, FAST_LOCK-ADAPTIVE_WAIT, USE_DNS_CACHE, USE_DNS_FAILOVER, USE_NAPTR, USE_DST_BLOCKLIST, HAVE_RESOLV_RES, TLS_PTHREAD_MUTEX_SHARED ADAPTIVE_WAIT_LOOPS 1024, MAX_RECV_BUFFER_SIZE 262144, MAX_URI_SIZE 1024, BUF_SIZE 65535, DEFAULT PKG_SIZE 8MB poll method support: poll, epoll_lt, epoll_et, sigio_rt, select. id: 272fdd compiled on 14:21:35 Oct 21 2024 with gcc 11.4.0 # dpkg -l |grep openssl ii openssl 3.0.2-0ubuntu1.18 amd64 Secure Sockets Layer toolkit - cryptographic utility Regards.

3 3

children monitoring
by Sergey Safarov 06 Jan '25

06 Jan '25

If TCP children are busy, Please correct me if I am wrong if the remote IP is not responding via TCP protocol, this leads to a timeout and this child is busy. If many remote endpoints are not reachable via TCP this can lead to the case when half or more children are by timeout operations. I think required to monitor how many children are busy. Is it possible to get such info using kamcmd utility, statistics module, or xprom module. Sergey

4 7

Crash on dialog_publish.
by satyaprakash ch 06 Jan '25

06 Jan '25

Hi, The kamailio version we are using is "5.7.4", We are facing a kamailio crash recently and it has generated the core-dump. Below is the backtrace for the coredump. Could you please assist with this crash. *Bachtrace:* #0 0x00007fa6b051748a in dialog_publish (state=0x7fa6b0519505 "terminated", ruri=0x7069734030783532, entity=0x7ffcdfb16510, peer=0x7ffcdfb16520, callid=0x7fa5b0527cb8, initiator=1, lifetime=10, localtag=0x0, remotetag=0x0, localtarget=0x7fa5b0527ce8, remotetarget=0x7ffcdfb16500, do_pubruri_localcheck=0, uuid=0x7fa5b0527d18) at dialog_publish.c:302 body = 0x0 uri = {s = 0x0, len = 0} publ = 0x0 size = 0 content_type = {s = 0x0, len = -1288318368} ruri_uri = {user = {s = 0x0, len = 0}, passwd = {s = 0x0, len = 0}, host = {s = 0x0, len = 0}, port = {s = 0x0, len = 0}, params = {s = 0x0, len = 0}, sip_params = {s = 0x0, len = 0}, headers = {s = 0x0, len = 0}, port_no = 0, proto = 0, type = ERROR_URI_T, flags = (unknown: 0), transport = {s = 0x0, len = 0}, ttl = {s = 0x0, len = 0}, user_param = {s = 0x0, len = 0}, maddr = {s = 0x0, len = 0}, method = {s = 0x0, len = 0}, lr = {s = 0x0, len = 0}, r2 = {s = 0x0, len = 0}, gr = {s = 0x0, len = 0}, transport_val = {s = 0x0, len = 0}, ttl_val = {s = 0x0, len = 0}, user_param_val = {s = 0x0, len = 0}, maddr_val = {s = 0x0, len = 0}, method_val = {s = 0x0, len = 0}, lr_val = {s = 0x0, len = 0}, r2_val = {s = 0x0, len = 0}, gr_val = {s = 0x0, len = 0}} __func__ = "dialog_publish" #1 0x00007fa6b0518f89 in dialog_publish_multi (state=0x7fa6b0519505 "terminated", ruris=0x7069734030783532, entity=0x7ffcdfb16510, peer=0x7ffcdfb16520, callid=0x7fa5b0527cb8, initiator=1, lifetime=10, localtag=0x0, remotetag=0x0, localtarget=0x7fa5b0527ce8, remotetarget=0x7ffcdfb16500, do_pubruri_localcheck=0, uuid=0x7fa5b0527d18) at dialog_publish.c:412 __func__ = "dialog_publish_multi" #2 0x00007fa6b0506944 in __dialog_sendpublish (dlg=0x7fa5b046a850, type=64, _params=0x7fa6b0437620 <params>) at pua_dialoginfo.c:402 tag = {s = 0x0, len = 0} uri = { s = 0x7fa5b0527d41 "sip:+1248XXXXXXX@10.XXX.X.Xm6ka0542f9v0lmjjc4npkamailio954739148rdb2847sip :XXXXXX.com:5060sip:521XXXx0;tgrp=51;trunk-context=198.X.XXX.XXX@10.XXX.XX.XX:5060padi-6761ad46-154e-1560padi\300\300\300\300", len = 29} identity_local = { s = 0x7fa5b0527d28 "sip:413XXXx0@10.XXX.X.XXsip :+1248XXXXXXXX@10.XXX.X.XXm6ka0542f9v0lmjjc4npkamailio954739148rdb2847sip :XXXXXX.com:5060sip:415XXXx0;tgrp=51;trunk-context=198.X.XXX.XXX@10.XXX.XX.XX:5060padi-6761ad46-154e-"..., len = 25} target = {s = 0x0, len = 0} dlginfo = 0x7fa5b0527c90 request = 0x7fa6b3361b58 __func__ = "__dialog_sendpublish" Thanks.

1 2

record_route() adding wrong sips URI when transport is missing (defaulting to UDP)
by Benoit Panizzon 06 Jan '25

06 Jan '25

Hi Gang A quick search via google did not reveal any useful hits. Kamailio 5.7.6 CPE (TLS) => SBC (UDP) => Registrar (UDP) => Core CPE uses a sips contact URI to register, no transport attribute! sips:315996608@172.29.255.11:5061 SBC uses Path functionality. Situation: Call to CPE I see the registrar adding this Record-Route: Record-Route: <sips:registrar-ip;lr;ftag=6159862378> The SBC is adding a r2=on Record-Route to indicate the change between UDP and TLS. The Route-Set received by the CPE then is: INVITE sips:315996608@172.29.255.11:5061 SIP/2.0 Record-Route: <sips:sbc-ip:5061;transport=tls;r2=on;lr;ftag=6159862378> Record-Route: <sips:sbc-ip;r2=on;lr;ftag=6159862378> Record-Route: <sips:registrar-ip;lr;ftag=6159862378> Record-Route: <sip:core-ip;lr;ftag=6159862378> 180 Ringing req 100rel is making it back as the Via header look right. But when the remote side is sending a PRACK, is has to follow the Routeset which if I am not mistaking, is not making it back because because the CORE does not know how to send this via TLS to the Registrar. Core logging: prepare_new_uac(): can't fwd to af 2, proto 3 (no corresponding listening socket) IMHO already the 2nd Record-Route added by the Registrar should not contain a sips: URI but a sip: URI (reflecting the transport attribute not being present, defaulting to UDP) Is this a known issue? Mit freundlichen Grüssen -Benoît Panizzon- -- I m p r o W a r e A G - Leiter Commerce Kunden ______________________________________________________ Zurlindenstrasse 29 Tel +41 61 826 93 00 CH-4133 Pratteln Fax +41 61 826 93 01 Schweiz Web http://www.imp.ch ______________________________________________________

3 4

Guidance Required for Kamailio Load Balancer Behavior in Attended Transfer Scenarios
by Kotappa Gandudi 03 Jan '25

03 Jan '25

Dear Kamailio Team, I hope this email finds you well. I am currently working on a project where Kamailio is being utilized as a load balancer to distribute calls across multiple FreeSWITCH instances. I am using the round-robin technique for load distribution, and it is working perfectly for regular call scenarios. However, I am encountering an issue when handling attended transfer calls. Specifically, for a normal call, Kamailio assigns one FreeSWITCH instance, but during the attended transfer call, it assigns another FreeSWITCH instance. This behavior disrupts the continuity of the call handling. My requirement is for Kamailio to use the same FreeSWITCH instance for the entire call, including any attended transfers. I believe maintaining the same FreeSWITCH instance for the entire call flow would ensure consistency and avoid potential issues. Could you please guide me on how to configure Kamailio to achieve this behavior? I would greatly appreciate any suggestions, configurations, or modules that can help resolve this issue. Thank you for your time and assistance. Looking forward to your response. Best regards, Kotappa Y Gandudi

4 4

how to call loose_route() after adding route header without using msg_apply_changes?
by Benoît Panizzon 31 Dec '24

31 Dec '24

Hi List To handle CPE which can only handle a limited number of route header (PacketCable SIP specification), I need to store and restore the route header (limited topology hiding only affecting Route Header) to present the CPE only one Route or RR header. I pass the ID of the stored route-set via RR param. So when I get a message back from the CPE is: * Take ID from RR param and pull stored route set from cache. * remove_hf("Route"); * insert_hf("Route: $avp(stored-routeset)\r\n"); * msg_apply_changes(); * loose_route(); I found loose_route() only to be able to correctly determine the next hop, if I first did apply the changes. But after adding rtpengine to the mix, I got back into the known situation that calling rtpengine (on the branch route) after calling msg_apply_changes() results in duplicate SDP. Is there a way to get loose_route() to operate on freshly added header without calling msg_apply_changes? -- Mit freundlichen Grüssen -Benoît Panizzon- @ HomeOffice und normal erreichbar -- I m p r o W a r e A G - Leiter Commerce Kunden ______________________________________________________ Zurlindenstrasse 29 Tel +41 61 826 93 00 CH-4133 Pratteln Fax +41 61 826 93 01 Schweiz Web http://www.imp.ch ______________________________________________________

8 28

Async HTTP request is sent but response is incorrectly processed
by Sergio Charrua 30 Dec '24

30 Dec '24

Hi all! still struggling with Async HTTP Client ... My current configuration is very simple: #!define HTTP_API_ROUTING_ENDPOINT_MAIN " http://MY_HTTP_REST_API:18085/Endpoint" include_file "listen.cfg" loadmodule "pv" loadmodule "tm" loadmodule "http_async_client" loadmodule "xlog" loadmodule "tmx" loadmodule "sl" loadmodule "textops" loadmodule "cfgutils" loadmodule "siputils" loadmodule "rr" loadmodule "ctl" loadmodule "jsonrpcs.so" loadmodule "jansson.so" modparam("pv", "shvset", "http_async=i:8"); modparam("http_async_client", "workers", 1); modparam("http_async_client", "connection_timeout", 500) modparam("http_async_client", "curl_verbose", 1) # for debug purpose #modparam("tm", "fr_timer", 750) #modparam("tm", "fr_inv_timer", 20000) #modparam("tm", "remap_503_500", 0) modparam("tm", "failure_reply_mode", 3) modparam("tm", "max_inv_lifetime",5000) # ----- jsonrpcs params ----- modparam("jsonrpcs", "pretty_format", 1) /* set the path to RPC fifo control file */ modparam("jsonrpcs", "fifo_name", "/tmp/kamailio_rpc.fifo") /* set the path to RPC unix socket control file */ modparam("jsonrpcs", "dgram_socket", "/tmp/kamailio_rpc.sock") debug=2 children=16 log_facility=LOG_LOCAL0 log_prefix="{$mt $hdr(CSeq) $ci} " disable_sctp = yes force_rport = yes rundir="/tmp" request_route { $avp(REQUEST_URL) = HTTP_API_ROUTING_ENDPOINT_MAIN; if ( is_method("ACK") ) { if ( t_check_trans() ) { t_relay(); } exit; } if(is_method("OPTIONS")){ sl_reply("200", "OK"); exit; } if(is_method("CANCEL")){ sl_reply("200","OK"); sl_reply("487","Request Terminated"); exit; } jansson_set("string","from" , $hdr(From) , "$var(req)"); jansson_set("string","to" , $hdr(To) , "$var(req)"); jansson_set("string","r-uri" , $ru , "$var(req)"); jansson_set("string","contact" , $hdr(Contact) , "$var(req)"); jansson_set("string","call-id" , $ci , "$var(req)"); xinfo("MAIN-ROUTE - Setting From=$hdr(From) To=$hdr(To) Contact=$hdr(Contact) ci=$ci \n"); if ( is_present_hf("Identity") ) jansson_set("string","identity" , $hdr(Identity) , "$var(req)"); if ( is_present_hf("P-Identity-Bypass") ) jansson_set("string","p-identity-bypass" , $hdr(P-Identity-Bypass) , "$var(req)"); if ( is_present_hf("P-Asserted-Identity") ) jansson_set("string","p-asserted-identity" , $hdr(P-Asserted-Identity) , "$var(req)"); if ( is_present_hf("P-STSH-UC") ) jansson_set("string","p-stsh-uc" , $hdr(P-STSH-UC) , "$var(req)"); jansson_set("string", "request-time" , $avp(requestTime), "$var(req)"); ## Create transaction before http_async_query; if (is_method("INVITE")){ t_newtran(); xinfo("MAIN-ROUTE - Sending request [$var(req)]\n"); route(HTTP_ASYNC_REQUEST); } exit; } route[HTTP_ASYNC_REQUEST] { xinfo("HTTP_ASYNC_REQUEST - Sending request [$var(req)]\n"); $http_req(all) = $null; $http_req(suspend) = 1; $http_req(timeout) = 1500; # 1 second timeout $http_req(method) = "POST"; $http_req(hdr) = "Content-Type: application/json"; $http_req(body) = $var(req); if ( !http_async_query($avp(REQUEST_URL), "HTTP_REPLY") ) { xinfo("HTTP_ASYNC_REQUEST - http_async_query $ci FAILED. Return is [$rc]\n"); }else{ send_reply("503", "Server Error"); exit; } } route[HTTP_REPLY] { xinfo("HTTP_REPLY - Xphere Response Code $http_rs - $ci\n"); if ( $http_ok && $http_rs==200) { xinfo("HTTP_REPLY - $ci reply ok\n"); jansson_xdecode($http_rb, "json"); xinfo("HTTP_REPLY - HTTP JSON Response: $http_rb \n"); if ( $xavp(json=>sip-response-code) == 300 ){ remove_hf("Contact"); append_to_reply("Contact: $xavp(json=>contact)\r\n"); ## "touch" the ruri to add the contact header for 3xx: $ru = $ru; send_reply("300", "Multiple Choice"); exit; }else{ xinfo("HTTP_REPLY - $ci Error Replied $xavp(json=>sip-response-code) - $xavp(json=>sip-response-text) \n"); send_reply("$xavp(json=>sip-response-code)","$xavp(json=>sip-response-text)"); exit; } } else { xinfo("HTTP_REPLY - HTTP Response Code $ci - $http_rs and error $http_err \n"); if (!$http_ok){ sl_reply("503", "No Connection to HTTP"); exit; }else if ($http_rs>=500){ sl_reply("503", "HTTP Processing Error"); exit; } else{ sl_reply("503", "Cannot Connect to HTTP"); exit; } } exit; } The configuration is based on Ben Kaufman's kamailio_http_async <https://github.com/whosgonna/kamailio_http_async> example, but I have extended it a bit. Currently all the requests are being sent to the HTTP REST API. I can't find queue/buffering issues... What I am experiencing is that , using SNGREP, I get this flow: 10.240.12.209:5060 10.242.17.146:5060 ----------*---------- ----------*---------- 17:33:46.689852 x INVITE (SDP) x +0.000530 x --------------------------> x 17:33:46.690382 x 100 trying -- your call is x +1.499822 x <-------------------------- x 17:33:48.190204 x 503 No Connection to HTTP x +0.000029 x <-------------------------- x 17:33:48.190233 x 500 I'm terribly sorry, se x +0.002130 x <-------------------------- x 17:33:48.192363 x ACK x +0.485836 x --------------------------> x 17:33:48.678199 x 500 I'm terribly sorry, se x +0.004881 x <<<------------------------ x 17:33:48.683080 x ACK x +0.973295 x ------------------------>>> x 17:33:49.656375 x 500 I'm terribly sorry, se x +0.004105 x <<<------------------------ x 17:33:49.660480 x ACK x +1.995420 x ------------------------>>> x 17:33:51.655900 x 500 I'm terribly sorry, se x +0.005361 x <<<-------------------------x 17:33:51.661261 x ACK x x ------------------------>>> x I suspected issues with the HTTP REST service, but when using tcpdump, I can find the Request from Kamailio and a reply back from the REST API service, so doesn't seem to be a problem on the HTTP REST service So my questions are: - why is Kamailio sending a "SIP 500 I'm terribly sorry..." error after a 503 No Connection to HTTP is sent and an "exit" command is executed? - how can I get rid of these messages? - Why Kamailio keeps sending responses back to SBC after the ACK when it is set to just "exit"? - Why is Kamailio not processing the response from HTTP REST API when it is correctly received? Logs are also weird: - I find a lot of these warnings: "Dec 29 17:38:34 ire-nprod-sip1 /usr/local/sbin/kamailio[590864]: WARNING: http_async_client [http_async_client_mod.c:529]: ah_get_status(): an async variable was read after http error, use $http_ok to check the request's status" and "Dec 29 17:38:34 ire-nprod-sip1 /usr/local/sbin/kamailio[590864]: ERROR: http_async_client [http_multi.c:621]: check_multi_info(): handle 0x7f375f786b30 returned error 7:" (I am still not using EVAPI module, that will be my next step, after the holiday season) (I suspect, however, that the REST API is not being able to handle the load in this test environment (between 400 to 600 CAPS), but that is another matter.) In attach, TCPDUMP, if anyone wishes to analyse. Atenciosamente / Kind Regards / Cordialement / Un saludo, *Sérgio Charrua*

3 5

http sync / async final tests
by Alexis Fidalgo 26 Dec '24

26 Dec '24

Just to share, and thanks for your valuable help. My conclusion after tests (on my scenario) is http backend: is a golang http server with a 250ms delay to answer, and answering back a json, running in 10 replicas inside a K8 with 4 nodes. test (from kamailio perspective): receive INVITE send 100 call the http, get the result, get a key from the result and set it as the value of Contact header send 302 receive ACK (discard on sync scenario because it’s stateless, use it in async to end transaction) Kamailio host: 1vcpu kamailio 3.8.4 bookworm running inside containerd (with podman) 16GB RAM sipp: run from the K8 nodes hosts (directly from OS) (everything is LAN connected on 2 different physical servers, hypervisor is VMWare i guess 7.0.3) using http_client (sync with stateless): 8 children -M8 -m64 no more than 60 calls/sec (15cps on each sipp instance) until sipp start to retransmit the INVITES using async_http_client: 8 children -M8 -m2048 it’s been ~36 hours now at 250cps each sipp instance with 0 errors 0 retransmission and the kamailio host is userlab@sblb1:~$ vmstat 1 procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 4 0 0 10285460 364684 5444240 0 0 0 22 36 68 0 0 100 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5284 6822 27 14 58 0 0 4 0 0 10285460 364684 5444240 0 0 0 0 5306 6752 29 15 57 0 0 3 0 0 10285460 364684 5444240 0 0 0 0 5304 6823 26 16 57 0 0 2 0 0 10285460 364684 5444240 0 0 0 12 5287 6761 26 16 57 0 0 2 0 0 10285460 364684 5444240 0 0 0 0 5337 6838 27 14 58 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5101 6724 25 17 59 0 0 3 0 0 10285460 364684 5444240 0 0 0 0 5066 6650 26 14 60 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5306 6829 23 17 59 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5219 6725 23 19 58 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5274 6770 26 13 61 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5295 6756 26 18 56 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5235 6786 22 16 62 0 0 1 0 0 10285460 364684 5444240 0 0 0 0 5315 6833 21 20 59 0 0 at least for my scenario, it worths to pay the price of the memory tradeoff. again, Ben, Alex, Sergio, all, thanks for your help. regards

1 0

help on how to get ACK
by Alexis Fidalgo 24 Dec '24

24 Dec '24

Hello all, moving just a bit aside of the http and async_http. After all the real useful and interesting thread on that topic what helped me, im facing a problem i cant deal with and need a hint at least. Scenario INVITE -> Kamailio on request_route ... if(is_method("INVITE")){ t_newtran(); http_async_query("http://nuc:8080", "HTTP_REPLY"); } … Kamailio -> 100 - Trying then route[HTTP_REPLY] { if ($http_ok) { xlog("L_INFO", "route[HTTP_REPLY]: status $http_rs\n"); xlog("L_INFO", "route[HTTP_REPLY]: body $http_rb\n"); t_reply(302,"Redirect"); } else { xlog("L_INFO", "route[HTTP_REPLY]: error $http_err)\n"); } } Kamailio -> 302 Redirect ACK -> Kamailio This last ACK, how can i read it and use it to terminate the transaction? because Kamailio keeps transmitting the 302 message 3 more times until the transaction is finished by a timer 42(44) DEBUG: tm [t_reply.c:1723]: t_retransmit_reply(): reply retransmitted. buf=0x7f4c44f9d680: SIP/2.0 3..., shmem=0x7f4c3fce7900: SIP/2.0 3 42(44) DEBUG: tm [t_reply.c:1723]: t_retransmit_reply(): reply retransmitted. buf=0x7f4c44f9d680: SIP/2.0 3..., shmem=0x7f4c3fce7900: SIP/2.0 3 42(44) DEBUG: tm [t_reply.c:1723]: t_retransmit_reply(): reply retransmitted. buf=0x7f4c44f9d680: SIP/2.0 3..., shmem=0x7f4c3fce7900: SIP/2.0 3 42(44) DEBUG: tm [timer.c:642]: wait_handler(): finished transaction: 0x7f4c3fcd35a0 (p:0x7f4c3fad85d0/n:0x7f4c3fad85d0) 42(44) DEBUG: tm [h_table.c:133]: free_cell_helper(): freeing transaction 0x7f4c3fcd35a0 from timer.c:651 in request_route i have if(is_method("ACK")){ if(!t_check_trans()){ t_release(); } exit; }

4 26

append_branch
by Alexis Fidalgo 24 Dec '24

24 Dec '24

Hello all, how can avoid to append_branch do this call to call SIP/2.0 302 Redirect Via: SIP/2.0/UDP 192.168.86.250:5060;branch=z9hG4bK-65665-5-0 From: sipp <sip:sipp@192.168.86.250:5060>;tag=65665SIPpTag005 To: 123456 <sip:123456@192.168.86.128:5060>;tag=a6a1c5f60faecf035a1ae5b6e96e979a-deb8da2e Call-ID: 5-65665(a)192.168.86.250 CSeq: 1 INVITE Contact: <sip:4093823670@otherdomain.com>, <sip:8888576523@otherdomain.com>, <sip:9719533197@otherdomain.com>, <sip:5302579529@otherdomain.com>, <sip:4972010218@otherdomain.com> Server: kamailio (5.8.4 (x86_64/linux)) Content-Length: 0 What i mean with “avoid”, call to call, contact should have only 1 branch, no a stack of previous. thanks

2 1

Transactions help
by Alexis Fidalgo 24 Dec '24

24 Dec '24

Hello, after the discussion regarding http and http_async, I started some labs to make a comparison and see what’s the trade off and the associated cost. Main problem (to me) is im not clear on how transactions behave. Im running some fast tests now before I start to deploy the final scenarios and 1. A new transaction is created before the http_async call (as the module example shows for http_async_query function 2. The http_async_query call is executed, receives a response and the HTTP_REPLY route for http_async_query is executed. 3. As we need to redirect the call, im adding Contact information with append_to_reply and then calling t_reply(302,”Redirect) Up to here INVITE is received, a 100 is answered and then the 302 with the expected Contact information (collected and built from the http_async_query response), problem is that the 302 is retransmitted 3 more times. So, im not even sure that im doing the correct actions to process the call this way. Is there a document/book/reference I can read to understand transactions better? By now I can not handle the call to be processed correctly (without the retransmissions) and finish the transaction the correct way. Any help, link, guide, will be appreciated. Regards.

6 17

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

sr-users