sr-users September 2023

sr-users@lists.kamailio.org

52 participants
58 discussions

Iterating Through Multiple IP Peers for DID Call Termination in Kamailio

by Shah Hussain Khattak

Hello Team, I hope you're all doing well. I'm reaching out to request your insights on the following scenario. Overview: I have successfully configured Kamailio to authenticate IP addresses, primarily for PBXs and customers unable to register. With this setup, I can flawlessly receive incoming traffic and forward it to our internal FreeSwitch servers. Current issue: Now, I'm focusing on managing traffic in the opposite direction—specifically for DIDs. I have a DID table set up in Kamailio that associates DID numbers with their respective IP peers. Upon receiving a call, a database lookup is performed to find the destination number, after which the call is terminated at the customer's IP. I'm facing difficulty in handling situations where a customer has multiple IP peers (for redundancy). My aim is to try each of these IPs sequentially when terminating a specific DID call towards them. The current logic I'm employing for the INVITE lookup looks like this: # Database lookup for INVITEs if (is_method("INVITE")) { sql_query("didrouting_db", "SELECT route_to FROM did_routing WHERE did_number='$rU'", "result"); if ($dbr(result=>rows) > 0) { $var(route_to) = $dbr(result=>[0,0]); xlog("L_INFO", "Routing DID $rU to $var(route_to)\n"); $du = "sip:" + $var(route_to) + ":5060"; } else { $du = "sip:" + FS_IP + ":5060"; } } The above logic works fine for a single IP. I intend to use a comma-separated list of IPs in my route_to column (for example 192.168.1.2,192.168.1.3, ...) and then iterate through these IPs one by one. I'm having trouble coming up with a workable logic or loop to achieve this. Would you be able to offer any guidance or suggestions? Please also recommend if there is a better approach available to handle this scenario. Thank you in advance for your valuable input. Regards, Shah Hussain

7 months, 2 weeks

TM max_inv_lifetime being reset by downstream responses

by James Browne

Hi all I see this old thread from 2015. - https://kamailio.org/mailman3/hyperkitty/list/sr-users@lists.kamailio.org/t… It's about max_inv_lifetime not working properly. I have this problem today: SIP-180 responses from downstream apparently reset max_inv_lifetime, and the transaction can live for hours (apparently indefinitely) even though max_inv_lifetime is set to just over a minute. By spacing the SIP-180 responses (from downstream) slightly longer apart than than max_inv_lifetime in kamailio, I don't get the problem, so it looks to me that this is TM not matching what's in the documentation (which mentions that this should be a hard limit). With this configuration, SIP-180 responses every minute result in the transaction never ending (until the client CANCELs the call). - kamailio.cfg:modparam("tm", "fr_timer", 3000) - kamailio.cfg:modparam("tm", "fr_inv_timer", 300000) - kamailio.cfg:modparam("tm", "max_inv_lifetime", 65000) - kamailio.cfg:modparam("tm", "noisy_ctimer", 1) That other thread ends with Daniel mentioning that he'd have to look at the code, but nothing further. Running "git log --grep max_inv_lifetime" shows me that the last/only update that wasn't just documentation was in 2007. Has anyone got an ideas? James

7 months, 2 weeks

Modifying SDP as drop-in replacement for overloaded Asterisk box - looking for help/paid consulting fast

by Markus

Hi list, I'm trying to use Kamailio 4.4.4 with rtpengine in a self-inflicted emergency situation (didn't monitor traffic growth properly and now encountering packet loss during peak times) as a drop-in replacement for an overloaded Asterisk box in a call-termination-to-upstream-carrier scenario. My test scenario is to make a call from a SIP softphone to Asterisk IP 1.1.1.1 -> Kamailio/rtpengine IP 2.2.2.2 -> Upstream carrier 3.3.3.3 sngrep on Kamailio box 2.2.2.2 - the following SDP will not work - carrier is rejecting it. Carrier is authenticating our calls based on our IP address 2.2.2.2, no username/pass involved. 2023/09/22 02:06:49.216136 2.2.2.2:5060 -> 3.3.3.3:5060 INVITE sip:+32xxxxxxxx@2.2.2.2;user=phone SIP/2.0 Record-Route: <sip:2.2.2.2;lr> Via: SIP/2.0/UDP 2.2.2.2;branch=z9hG4bKd9c3.d6fa3abe5d52b827e2054de5573028e0.0 Via: SIP/2.0/UDP 1.1.1.1:5060;branch=z9hG4bK473270e8 Max-Forwards: 69 From: "61xxxxxxxxx" <sip:+61xxxxxxxxx@1.1.1.1>;tag=as3d75aadd To: <sip:+32xxxxxxxx@2.2.2.2;user=phone> Contact: <sip:+61xxxxxxxxx@1.1.1.1:5060> Call-ID: 3f31e1622a72b6d17f24e42362f4f1d0@1.1.1.1:5060 CSeq: 102 INVITE User-Agent: Asterisk PBX 20.0.0 Date: Fri, 22 Sep 2023 00:06:50 GMT Session-Expires: 1800 Min-SE: 90 Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO, PUBLISH, MESSAGE Supported: replaces, timer P-Asserted-Identity: <sip:+61xxxxxxxxx@2.2.2.2;user=phone> Content-Type: application/sdp Content-Length: 314 X-SIP: 1.1.1.1 v=0 o=root 1093000903 1093000903 IN IP4 1.1.1.1 s=Asterisk PBX 20.0.0 c=IN IP4 2.2.2.2 t=0 0 m=audio 25742 RTP/AVP 8 9 0 101 a=maxptime:150 a=rtpmap:8 PCMA/8000 a=rtpmap:9 G722/8000 a=rtpmap:0 PCMU/8000 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-16 a=sendrecv a=rtcp:25743 a=ptime:20 I'm comparing this rejected INVITE to a successful INVITE sent by the original Asterisk box at IP 2.2.2.2 (now Kamailio box) to the carrier without Kamailio in the path, and these are the differences I noticed, and probably the things I have to mimick with Kamailio in order to make it work: INVITE sip:+32xxxxxxxxx@2.2.2.2;user=phone SIP/2.0 should be INVITE sip:+32xxxxxxxxx@3.3.3.3;user=phone SIP/2.0 To: <sip:+32xxxxxxxx@2.2.2.2;user=phone> should be To: <sip:+32xxxxxxxx@3.3.3.3;user=phone> From: "61xxxxxxxxx" <sip:+61xxxxxxxxx@1.1.1.1>;tag=as3d75aadd should be From: "61xxxxxxxxx" <sip:+61xxxxxxxxx@2.2.2.2>;tag=as3d75aadd Contact: <sip:+61xxxxxxxxx@1.1.1.1:5060> should be Contact: <sip:+61xxxxxxxxx@2.2.2.2:5060> Call-ID: 3f31e1622a72b6d17f24e42362f4f1d0@1.1.1.1:5060 should be Call-ID: 3f31e1622a72b6d17f24e42362f4f1d0@2.2.2.2:5060 o=root 1093000903 1093000903 IN IP4 1.1.1.1 should be o=root 1093000903 1093000903 IN IP4 2.2.2.2 My kamailio.cfg can be found here: https://pastebin.com/6PKcRjPU These are the Asterisk boxes I want to originate calls from to Kamailio: [root@voip30 ~]# kamctl address show +-----+-----+----------+------+------+-----------+ | id | grp | ip_addr | mask | port | tag | +-----+-----+----------+------+------+-----------+ | 195 | 1 | 1.1.1.1 | 32 | 0 | voip20.sv | | 196 | 1 | 1.1.1.2 | 32 | 0 | voip21.sv | | 197 | 1 | 1.1.1.3 | 32 | 0 | voip22.sv | | 198 | 1 | 1.1.1.4 | 32 | 0 | voip23.sv | | 199 | 1 | 1.1.1.5 | 32 | 0 | voip24.sv | | 200 | 1 | 1.1.1.6 | 32 | 0 | voip25.sv | | 201 | 1 | 1.1.1.7 | 32 | 0 | voip26.sv | | 202 | 1 | 1.1.1.8 | 32 | 0 | voip27.sv | | 203 | 1 | 1.1.1.9 | 32 | 0 | voip28.sv | +-----+-----+----------+------+------+-----------+ This is the upstream carrier I want Kamailio to proxy calls to: [root@voip30 ~]# kamctl dispatcher show dispatcher gateways +----+-------+------------------+-------+-------+------------+------+ | id | setid | destination | flags | prio. | attrs | desc | +----+-------+------------------+-------+-------+------------+------+ | 12 | 1 | sip:3.3.3.3:5060 | 0 | 0 | weight=100 | | +----+-------+------------------+-------+-------+------------+------+ (output manually slightly modified to look properly over E-Mail) As you might have guessed I'm a Kamailio noob... and don't have the resources to learn it as fast as I must to avoid further packet loss. If there's anyone available who can help me to get this done today, optionally in exchange for money, I'd be grateful. Thank you! Markus

7 months, 2 weeks

Multi-dimensional data in htable

by Ben Kaufman

What's the best way to store multi-dimensional data within an htable? For example, storing an avp stack: $avp(foo) = "first"; $avp(foo) = "second"; $sht(bar=>foo) = $avp(foo); The result of which is that only $avp(foo[0]) is stored in the htable: kamcmd htable.dump bar { entry: 2 size: 1 slot: { { name: bar value: second type: str } } } The htable documentation shows support for lists, but it doesn't make clear if this is really usable outside of loading from a database, and looks like it's short hand for managing/creating linked lists. The other option would be to simply serialize the data before storing it in the htable - which is fine, I just want to be sure I'm not overlooking a more convenient method. Regards, Kaufman

7 months, 2 weeks

no corresponding socket found error

by David Escartín

Hello all we are seeing these kind of logs in a Debian GNU/Linux 11 using kamailio 5.5.6 Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155702]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:164.152.22.248:5060) Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155698]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.184:5060) Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155695]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.181:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155693]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:38.102.250.60:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155707]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:87.1.1.27:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155707]: ERROR: tm [ut.h:302]: uri2dst2(): no corresponding socket for "87.1.1.27" af 2 Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155707]: ERROR: tm [t_fwd.c:470]: prepare_new_uac(): can't fwd to af 2, proto 1 (no corresponding listening socket) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155697]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.184:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155694]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:192.40.216.97:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155694]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.180:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155714]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:87.1.1.27:5060) The kamailio instance we are using is receiving around 1500 calls per second in average when we do start seeing these errors more frequently In this instance we are using multihomed the listen address list is children=14 socket_workers=1 listen=udp:192.168.99.70:5081 listen=udp:192.168.99.81:5060 listen=udp:87.1.1.27:5060 listen=tcp:192.168.96.105:8095 listen=tcp:87.1.1.27:5060 tcp_children=6 port=5060 We use the first worker only to perform the OPTIONS requests from the dispatcher module. The sockets listen=udp:192.168.99.81:5060 listen=udp:87.1.1.27:5060 are used to communicate both A and B legs so when a message is received at 192.168.99.81:5060 we send it to 87.1.1.27:5060 and viceversa We were using $fs variable before doing the t_relay() function. This way is working, but when load increases, seems sometimes kamailio doesn't get the proper socket to forward the reply. I think the errors mainly are related to responses which are being forwarded, like if the function get_sock_info_list was not able to retrieve the listen interfaces we have tried to set in the onreply routes the commands set_recv_socket("udp:192.168.99.81:5060"); set_send_socket("udp:87.1.1.27:5060"); (when reply goes from private to public domain) and it seems it reduces the number of Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155702]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:164.152.22.248:5060) Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155698]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.184:5060) Sep 21 17:41:13 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155695]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:208.74.138.181:5060) Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155693]: ERROR: <core> [core/forward.c:183]: get_out_socket(): no corresponding socket found for(udp:38.102.250.60:5060) errors But i honestly don't know why the Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155707]: ERROR: tm [ut.h:302]: uri2dst2(): no corresponding socket for "87.1.1.27" af 2 Sep 21 17:41:14 lax-dedge-1 /usr/local/kamailio/sbin/kamailio[3155707]: ERROR: tm [t_fwd.c:470]: prepare_new_uac(): can't fwd to af 2, proto 1 (no corresponding listening socket) appear We aldso tried to increase the kernel buffers rmem and wmem from 208KB (as we had by default) to 4MB Do you know a reason which could cause these logs to appear? Any setup that could mitigate them? thanks a lot and regards david escartin -- <http://www.sonoc.io/> David Escartín NOC engineer www.sonoc.io [image: LinkedIn] <https://www.linkedin.com/company/sonoc> [image: Twitter] <https://twitter.com/sonoc_> [image: Instagram] <https://www.instagram.com/sonoc.io/> This e-mail is for the exclusive use of its recipients and may contain business secrets or other confidential or privileged information. Any unauthorised use, copying, publication or distribution of this e-mail is strictly prohibited. If you are not the intended recipient, please inform us immediately by replying to this e-mail and delete it, including any attachments or copies on your system. In accordance with the GDPR (EU) 2016/679 and the LOPDGDD 3/2018, we inform you that this e-mail address and/or any other personal data you have provided us with will be treated by SONOC with absolute confidentiality and with the only purpose of providing you with the requested services, due to your condition as a client, supplier or because you have requested information from us at any time. These data will only be kept for as long as required to comply with legal obligations. You can exercise your rights at any time by sending an e-mail to: dataprotection(a)sonoc.io. <dataprotection(a)sonoc.io>

7 months, 2 weeks

"no corresponding socket found" logs

by David Escartín

7 months, 2 weeks

Change in mailing list configuration related to DMARC spam filtering

by Henning Westerholt

Hello, due to some changes from certain cloud infrastructure provider related to spam filtering with DMARC, we need to change the list manager "mailman" configuration for the following lists: * sr-users * sr-user-es * sr-dev * business * sr-docs * kamailio-announce * the administrative mailing list The change will impact the "From" that is displayed in your email program. Without that change large e-mail providers (Gmail, Microsoft O365) will reject messages from a sender with certain strict DMARC policy. This is already affecting several people on the lists (for example with GitHub notice mails), but over time the problem will certainly increase. Regarding the technical change: We will munge the From: header so it doesn't contain the domain that triggers the DMARC rejection. Essentially, the Mailman list takes ownership of the message and injects its own address into the From: header. This can affect reply-to-sender, although we add the original From: address in the Reply-To: header (or sometimes the Cc: header) to reduce the impact of this. We plan to do this for all emails unconditionally, regardless of their sender DMARC policy, to have an identical behaviour for all users. Please refer to this [1] page for more details. For a general description of DMARC refer to this page [2]. This change should have not a big impact for you, but some users might need to adapt some mail filters. Therefore, we would like to gather feedback on that change from you. We expect that the most visible change is that It will be not possible anymore to just reply by e-mail to mails from GitHub and have them "automatically" added to the discussed issue. This feature was so far used rarely and only from a few people. We plan to implement this change by the end of this week, when its done it will be also shortly announced. Best regards, Henning Westerholt [1] https://wiki.list.org/DEV/DMARC [2] https://dmarc.org/overview/ -- Henning Westerholt - https://skalatan.de/blog/ Kamailio services - https://gilawa.com<https://gilawa.com/>

7 months, 2 weeks

RTP traffic through kamailio

by Shravan

7 months, 2 weeks

insert_dlg_timer(): Trying to insert a bogus dlg

by Benoît Panizzon

Hi Team Could anyone point me in the direction where I could find out more what's going wrong and what could be the cause of messages like: Sep 22 07:32:29 dev-core02 kamailio[915066]: CRITICAL: dialog [dlg_timer.c:129]: insert_dlg_timer(): Trying to insert a bogus dlg tl=0x7f9617d967b0 tl->next=0x7f9618e11198 tl->prev=0x7f9617e20548 Sep 22 07:32:29 dev-core02 kamailio[915066]: CRITICAL: dialog [dlg_dmq.c:314]: dlg_dmq_handle_msg(): Unable to insert dlg timer 0x7f9617d96750 [1598:12168] Are the numbers in brackets [hash_id:hash_entry] ? -- 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 ______________________________________________________

7 months, 2 weeks

Dialog: bogus event 2 in state 3

by Benoît Panizzon

Hi Team I'm still hunting dialog errors which cause my dialog counters to divert and stick around, basically exploding in my face now and then. I suppose they are DMQ related (maybe packet loss, maybe retransmissions, maybe generic overloading of the virtualized kamailio instances) causing issues with inserting dialog timers on the peer etc. So for the moment I'm running only one instance, no DMQ peer. What I still occasionally see are messages: INVITE]dialog [dlg_hash.c:1182]: next_state_dlg(): bogus event 2 in state 3 for dlg 0x7ff906294be8 [2482:872] with clid 'A8459C74@7f33ff47' and tags '7f33ff47+1+7d820120+89494caf' '3904288054-1882699775' I found an ancient post, confirming this to be an open bug in Kamailio 1.4 and race condition between the 200 OK reply to an INVITE and the ACK confirming the 200 OK. Someone attempted to backport a fix for opensips to kamailio. Indeed, I see this message occur occasionally after managing the 200 OK reply and routing the ACK. Does anyone know, if this is still a open bug with Kamailio 5.5? Or what else could be the cause of this message? -- 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 ______________________________________________________

7 months, 2 weeks

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 September 2023