sr-users

sr-users@lists.kamailio.org

4 participants
37456 discussions

using kamailio to send invite to webrtc clients (with freeswitch)
by shwetank singh 09 May '23

09 May '23

Hello experts I have a setup where webrtc client registrations (from behind a NAT / VPN) are on kamailio and the INVITE to begin a call is sent by freeswitch to kamailio which is then expected to relay it to the subscribed webrtc client. i understand, from documentation and general thinking, that kamailio cannot open a new connection to such a client and should reuse an open connection (a websocket in my case). i have setup kamailio as generic as it may sound and expect t_relay() to do the magic of forwarding the invite, however, from logs i don't see anything indicating kamailio is using any existing connection and hence my outgoing invite requests are stuck showing pointed towards a vpn gateway which was the registered "location" of client (which is the public ip of VPN gateway) please advise what relevant information can i produce here to make the query more readable and clear. headed here after trying just about all the resources available on the topic. e.g. if my location record looks like AoR: 1004(a)mydomain.com Contacts: - Contact: Address: sip:1004@11.22.33.44:49179 CFlags: 64 CSeq: 3087 Call-ID: 189jss21e24ntve1ggsuli Expires: 593 Flags: 0 Instance: <urn:uuid:8a936bbc-514f-4755-bed0-edc644dd4453> KA-Roundtrip: 0 Keepalive: 1 Last-Keepalive: 1683652755 Last-Modified: 1683652755 Methods: 7071 Path: <sip:10.10.6.196:3040;received=sip:mydomain.com ;lr;nat=yes> Q: -1 Received: sip:11.22.33.44:49179;transport=ws Reg-Id: 1 Ruid: uloc-645601c4-131220-3764 Server-Id: 0 Socket: tls:1.2.3.4:4443 State: CS_SYNC Tcpconn-Id: 3916 User-Agent: JsSIP 3.9.0 i try to lookup the location and find the address, and then route to that address by setting $ru in sngrep it does show that kamailio is trying to send INVITE to 11.22.33.44:port (which is the VPN gateway) but it isn't exactly clear from logs if this is being done over an existing connection. how does kamailio identify if a connection exists for a user (for each contact in AOR, if multiple contacts exist for a user -- i.e. different browser sessions?) if(reg_fetch_contacts("location", "$ru", "caller")) { $var(i) = 0; while($var(i) < $(ulc(caller=>count))) { $ru = $(ulc(caller=>addr)[$var(i)]); $var(i) = $var(i) + 1; setflag(FLT_DLGINFO); dlg_manage(); t_relay(); } } -- Shwetank

3 2

Performing async_http calls on stateless Loadbalancers
by Julius Flohr 09 May '23

09 May '23

Hi List, we are running several stateless SIP Loadbalancers using Kamaillio 5.1 and 5.6. Now we would like to perform an async_http request if an INVITE matches certain criteria. The request itself is performed very similar to how it's described here [1] This uses the transaction module in order to assure that the response can be mapped to the original request. However, this also means that the processing of the entire request becomes stateful which we would like to avoid. We tried to use t_release() in the callback route (in [1] this route is called "RELAY_API_RESPONSE"). However, documentation says that this is only allowed in the REQUEST_ROUTE. Is there an easy way to circumnavigate this issue? Cheers [1]: https://wazo-platform.org/blog/kamailio-routing-with-rtjson-and-http-async-… -- Julius Flohr - Voice Engineer - flohr(a)sipgate.de Telefon: +49 (0)211-63 55 55-0 Mobil: +49 (0)1579-000-0205 Telefax: +49 (0)211-63 55 55-22 sipgate GmbH - Gladbacher Str. 74 - 40219 Düsseldorf HRB Düsseldorf 39841 - Geschäftsführer: Thilo Salmon, Tim Mois Steuernummer: 106/5724/7147, Umsatzsteuer-ID: DE219349391 www.sipgate.de - www.sipgate.co.uk

3 3

Upgrade to 5.6.4 => [core/mem/q_malloc.c:519]: qm_free(): BUG: freeing already freed pointer
by Benoît Panizzon 08 May '23

08 May '23

Hi team I upgraded our lab to 5.6.4 from the official repository for ubuntu Jammy. Now I get crashes related to TLS: May 5 14:58:40 dev-cpereg01 kamailio[36147]: CRITICAL: <core> [core/mem/q_malloc.c:519]: qm_free(): BUG: freeing already freed pointer (0x7f4fedbed4e8), called from tls: tls_init.c: ser_free(323), first free tls: tls_init.c: ser_f ree(323) - ignoring May 5 14:58:40 dev-cpereg01 kamailio[36147]: CRITICAL: <core> [core/mem/q_malloc.c:519]: qm_free(): BUG: freeing already freed pointer (0x7f4fedd72b88), called from tls: tls_init.c: ser_free(323), first free tls: tls_init.c: ser_f ree(323) - ignoring May 5 14:58:40 dev-cpereg01 kamailio[36177]: CRITICAL: <core> [core/mem/q_malloc.c:519]: qm_free(): BUG: freeing already freed pointer (0x7f4fedd9e4a0), called from tls: tls_init.c: ser_free(323), first free tls: tls_init.c: ser_f ree(323) - ignoring May 5 14:58:40 dev-cpereg01 kamailio[36157]: CRITICAL: <core> [core/mem/q_malloc.c:519]: qm_free(): BUG: freeing already freed pointer (0x7f4fedd9e428), called from tls: tls_init.c: ser_free(323), first free tls: tls_init.c: ser_f ree(323) - ignoring And a bit later on: May 5 14:58:46 dev-cpereg01 kamailio[36179]: CRITICAL: <core> [core/pass_fd.c:277]: receive_fd(): EOF on 34 May 5 14:58:46 dev-cpereg01 kamailio[36143]: ALERT: <core> [main.c:774]: handle_sigs(): child process 36161 exited by a signal 11 May 5 14:58:46 dev-cpereg01 kamailio[36143]: ALERT: <core> [main.c:777]: handle_sigs(): core was generated May 5 14:58:46 dev-cpereg01 kamailio[36143]: INFO: <core> [main.c:799]: handle_sigs(): terminating due to SIGCHLD Has there been a fundamental change in TLS? Or did I miss something important to do on a 5.5 => 5.6 upgrade? -- 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 ______________________________________________________

4 5

TLS1.2 ciphers
by Sergiu Pojoga 05 May '23

05 May '23

Hi gang, Trying to figure out why Kamailio won't negotiate specific TLS1.2 ciphers, such as for example TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 i.e. *ECDHE-RSA-AES256-GCM-SHA384* OpenSSL lists the cipher as supported, however Kamailio won't negotiate it: # nmap --script ssl-enum-ciphers -p 5061 mysbc.domain.com PORT STATE SERVICE 5061/tcp open sip-tls | ssl-enum-ciphers: | TLSv1.2: | ciphers: | TLS_RSA_WITH_AES_256_GCM_SHA384 (rsa 2048) - A | TLS_RSA_WITH_AES_256_CBC_SHA256 (rsa 2048) - A | TLS_RSA_WITH_AES_256_CBC_SHA (rsa 2048) - A | TLS_RSA_WITH_CAMELLIA_256_CBC_SHA (rsa 2048) - A | TLS_RSA_WITH_AES_128_GCM_SHA256 (rsa 2048) - A | TLS_RSA_WITH_AES_128_CBC_SHA256 (rsa 2048) - A | TLS_RSA_WITH_AES_128_CBC_SHA (rsa 2048) - A | TLS_RSA_WITH_SEED_CBC_SHA (rsa 2048) - A | TLS_RSA_WITH_CAMELLIA_128_CBC_SHA (rsa 2048) - A | TLS_RSA_WITH_RC4_128_SHA (rsa 2048) - C | TLS_RSA_WITH_RC4_128_MD5 (rsa 2048) - C | TLS_RSA_WITH_3DES_EDE_CBC_SHA (rsa 2048) - C | compressors: | NULL | cipher preference: server | warnings: | 64-bit block cipher 3DES vulnerable to SWEET32 attack | Broken cipher RC4 is deprecated by RFC 7465 | Ciphersuite uses MD5 for message integrity |_ least strength: C OS: - outdated Debian 8 Jessie - OpenSSL 1.0.1t 3 May 2016 - openssl ciphers -v | grep '*ECDHE-RSA-AES256-GCM-SHA384*' - ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 Kx=ECDH Au=RSA Enc=AESGCM(256) Mac=AEAD # kamailio -V version: kamailio 5.5.6 (x86_64/linux) ad1244 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 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. compiled on 21:08:20 Apr 13 2023 with gcc 4.9.2 modparam("tls", "cipher_list", "ALL") method = TLSv1.2+ Elliptic Curve Diffie-Hellman (EDCH) <https://kamailio.org/docs/modules/devel/modules/tls.html#tls.compile>-Ciphers are only supported in OpenSSL 1.0.0e and later. Any suggestions? Thanks.

5 5

Using a second Perl library
by David Cunningham 04 May '23

04 May '23

Hello, We are using a product which already uses the "app_perl" module and sets the "filename" parameter to its own Perl library. However, we want to add more Perl code without having to modify that product's library. Does anyone know of a way to do this? To have Kamailio use Perl code from two different places? I guess in the worst case scenario we could use "app_python3", and have some Python code call subprocess.run() on the Perl code. Thanks very much, -- David Cunningham, Voisonics Limited http://voisonics.com/ USA: +1 213 221 1092 New Zealand: +64 (0)28 2558 3782

1 0

Development is open for future release of v5.8.x series
by Daniel-Constantin Mierla 04 May '23

04 May '23

Hello, the branch 5.7 was created, therefore the master branch is open for adding new features, to be part of future release series v5.8.x (or whatever version is decided for next series). Any bug fix committed to master that applies to 5.7.x or older stable branches should be backported as usual with "git cherry-pick -x ..." to appropriate branches like 5.7 or 5.6. Expect that v5.7.0 will be released in a few weeks from now. Based on the workflow used during the past years, the next future release v5.8.0 should be out after another 8-10 months of development, plus 1-2 months of testing, so sometime during the first part of 2024. Cheers, Daniel -- Daniel-Constantin Mierla -- www.asipto.com www.twitter.com/miconda -- www.linkedin.com/in/miconda Kamailio World Conference - June 5-7, 2023 - www.kamailioworld.com

1 0

Branch 5.7 has been created
by Daniel-Constantin Mierla 04 May '23

04 May '23

Hello, the branch 5.7 has been created, to be used for releasing v5.7.x series. To check out this branch, the following commands can be used: git clone https://github.com/kamailio/kamailio kamailio-5.7 cd kamailio-5.7 git checkout -b 5.7 origin/5.7 Pushing commits in this branch: git push origin 5.7:5.7 Note that 5.7 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 5.7 branch: git cherry-pick -x COMMITID In few weeks, the first release from branch 5.7 will be out, respectively Kamailio v5.7.0. Cheers, Daniel -- Daniel-Constantin Mierla -- www.asipto.com www.twitter.com/miconda -- www.linkedin.com/in/miconda Kamailio World Conference - June 5-7, 2023 - www.kamailioworld.com

1 0

send reply from branch route?
by Benoit Panizzon 04 May '23

04 May '23

Hello Is there a possibility to send a reply from within a branch route? Experimental case (yes, I know it's a non working set-up, but I just wonder if that would be possible). Two registrar nodes, location information shared between the two. CPE registers via TLS, therefore the existing connection shall be used towards the CPE and that only exists on one of the registrars. So in the Branch Route I would like to do something like: $var(socket) = $(ulc(aor=>socket)[$T_branch_idx]); if ($var(socket) == 0) { send_reply("503","no local socket"); } This would cause the core routing instance to select the next registrar from the dispatcher list which hopefully holds the active socket. But none of the 'send_reply' variants seem to be allowed within a branch route. Or is there a better solution? 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 6

Branching 5.7 on git repo for 5.7.x releases
by Daniel-Constantin Mierla 04 May '23

04 May '23

Hello, I am planning to split 5.7.x series on its own git branch 5.7 on Thursday, May 4, 2023. After that, master branch will be open for new features (to become part of 5.8 or what ever version will be decided to be next). Usually testing should continue for branch 5.7 for another 1-2 weeks, then, if no relevant road blockers pop up, we should release 5.7.0 as first stable version in the 5.7.x series. Cheers, Daniel -- Daniel-Constantin Mierla -- www.asipto.com www.twitter.com/miconda -- www.linkedin.com/in/miconda Kamailio World Conference - June 5-7, 2023 - www.kamailioworld.com

1 1

uac.reg_add to database
by Jon Bonilla (Manwe) 04 May '23

04 May '23

Hi 5.6.4 I'm trying to add uacreg logic to my setup. When I add a new entry via rpc I see that the reg is stored in mem but not inserted in the database for persistency. My uac params are like this: modparam("uac", "reg_db_url",DBURL) modparam("uac", "reg_timer_interval", 900) modparam("uac", "reg_retry_interval", 10) modparam("uac", "reg_random_delay", 0) modparam("uac", "reg_hash_size", 12) ## 5.6 modparam("uac", "reg_db_table", "kam_uacreg") modparam("uac", "reg_contact_addr", "192.168.255.254:5060") #Won't be used #modparam("uac", "default_socket", "udp:192.168.0.125:5060") #Won't be used modparam("uac", "reg_keep_callid", 1) modparam("uac", "reg_active", 1) modparam("uac", "reg_gc_interval", 30) modparam("uac", "reg_use_domain", 0) I'm used to add via rpc and have db persistency. Like for exmaple using the domain module. Is this a bug in my config or does this module behave different? Should I insert in the db and call to enable via RPC? If so, what would be the best way to enable only the new inserted register instead of reloading all the regs? It's expected to have many changes and do not want to have full reloads all the time. cheers, Jon -- PekePBX, the multitenant PBX solution https://pekepbx.com

1 0

Jump to page:

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