June 2019 - sr-users - kamailio.org

Kamailio stop to process incoming SIP traffic via TCP.
by Kristijan Vrban 16 Jan '20

16 Jan '20

Hi kamailios, i have a creepy situation with v5.2.1 stable Kamilio. After a day or so, Kamailio stop to process incoming SIP traffic via TCP. The incoming TCP network packages get TCP-ACK from the OS (Debian 9, 4.18.0-15-generic-Linux) but Kamailio does not show any processing for the SIP-Traffic incoming via TCP. No logs, nothing. While traffic via UDP is working just totally fine. When i look via command "netstat -ntp" is see, that the Recv-Q get bigger and bigger. e.g.: Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name tcp 4566 0 172.17.217.12:5060 xxx.xxx.xxx.xxx:57252 ESTABLISHED 31347/kamailio After Kamailio restart, all is working fine again for a day. We have maybe 10-20 devices online via TCP and low call volume (1-2 call per minute). The only settings for tcp we have is "tcp_delayed_ack=no" How to could we debug this situation? Again, no error, no warings in the log. Just nothing. Kristijan

11 46

WebRTC ACK Protocol
by Ilie Soltanici 08 Jan '20

08 Jan '20

Hello, I am trying to set up a WebRTC2SIP Gateway by using Kamailio and rtpengine. So far, everything is working fine, I'm able to register an extension and make a call, but for some reason, when i'm trying to call a WebRTC extension from any SIP Extension Kamailio is sending INVITE, WebRTC extension is sending back 200 OK, and then Kamailio is trying to send an ACK through UDP protocol, and not through wss, as it's supposed to do. This is how invite is looking: INVITE sip:nl7oe4ss@vjbh7r4im6j7.invalid;transport=wss SIP/2.0 Record-Route: <sip:my-company.net ;transport=udp;ftag=as1789445c;lr=on;nat=yes> Via: SIP/2.0/WSS 123.123.123.123:10443 ;branch=z9hG4bKe655.29d7c135a302f3eb803902d4f5a8da7e.0 Via: SIP/2.0/UDP 192.168.50.237:5060 ;received=192.168.50.237;branch=z9hG4bK7d2e534e;rport=5060 Max-Forwards: 70 From: "WebRTC" <sip:11@my-company.net>;tag=as1789445c To: <sip:15@192.168.50.210:5060> Contact: <sip:11@192.168.50.237:5060> Call-ID: 7fc800de060197fa2315c93763873092(a)my-company.net CSeq: 102 INVITE User-Agent: Proxy Date: Wed, 03 Apr 2019 17:11:41 GMT Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY, INFO, PUBLISH, MESSAGE Supported: replaces, timer Alert-Info: Content-Type: application/sdp Content-Length: 596 Server: SIP Proxy and then WebRTC app is replying with 200 OK: SIP/2.0 200 OK Record-Route: <sip:my-company.net ;transport=udp;ftag=as1789445c;lr=on;nat=yes> Via: SIP/2.0/WSS 123.123.123.123:10443 ;branch=z9hG4bKe655.29d7c135a302f3eb803902d4f5a8da7e.0 Via: SIP/2.0/UDP 192.168.50.237:5060 ;received=192.168.50.237;branch=z9hG4bK7d2e534e;rport=5060 To: <sip:15@192.168.50.210:5060>;tag=dk4fa8ftt6 From: "WebRTC" <sip:11@my-company.net>;tag=as1789445c Call-ID: 7fc800de060197fa2315c93763873092(a)my-company.net CSeq: 102 INVITE Contact: <sip:nl7oe4ss@vjbh7r4im6j7.invalid;transport=wss> Allow: ACK,CANCEL,INVITE,MESSAGE,BYE,OPTIONS,INFO,NOTIFY,REFER Supported: outbound User-Agent: Proxy-WEBRTC Content-Type: application/sdp Content-Length: 901 and finally, Kamailio is trying to send this ack through UDP protocol: ACK sip:nl7oe4ss@22.22.22.22:57421;transport=wss SIP/2.0 Via: SIP/2.0/UDP 192.168.50.237:5060;branch=z9hG4bK56363ddf;rport Route: <sip:my-company.net;transport=udp;ftag=as1789445c;lr=on;nat=yes> Max-Forwards: 70 From: "WebRTC" <sip:11@my-company.net>;tag=as1789445c To: <sip:15@192.168.50.210:5060>;tag=dk4fa8ftt6 Contact: <sip:11@192.168.50.237:5060> Call-ID: 7fc800de060197fa2315c93763873092(a)my-company.net CSeq: 102 ACK User-Agent: Proxy Content-Length: 0 If i'm trying to force it through TLS, i'm receiving error: get_send_socket2(): protocol/port mismatch (forced tls:123.123.123.123:10443, to udp:22.22.22.22:23317) Can someone point me in the right direction, please? Thank you.

3 4

rtpproxy does not response, disable it
by Zodiac 03 Jan '20

03 Jan '20

Dear friends, I am working on a program on Kamailio and rtpengine proxy. I am wondering whether can I set Kamailio and rtpengine daemon on different physical machines. For example, I set Kamailio on a machine with IP address:10.109.247.80, and launch rtpengine daemon on another machine with interface parameter as 10.109.247.90 and ng port 7723. I set parameter in Kamailio.cfg with modparam(“rtpengine”, “rtpengine_sock”, “udp:10.109.247.90:7723”). Unfortunately I got debug message like this: ERROR: rtpengine [rtpengine.c:1710]: send_rtpp_command(): can't send command to a RTP proxy ERROR: rtpengine [rtpengine.c:1746]: send_rtpp_command(): proxy <udp:10.109.247.90:7723> does not respond, disable it ERROR: rtpengine [rtpengine.c:1616]: rtpp_test(): proxy did not respond to ping And, I also tried to set Kamailio and rtpengine daemon in a same machine,and use modparam(“rtpengine”, “rtpengine_sock”, “udp:localhost:7723”). And Kamailio can work functionally under this situation. rtpengine daemon can receive ping message from Kamailio and rtpengine daemon can work as suspected. So for the later case, is it supposed that Kamailio be in the same machine with same localhost address? Otherwise, what’s the reason for my ERROR? ------------------------------------ 北京邮电大学网络技术研究院网络与交换技术国家重点实验室田军 +86 18810315790 mozillafire(a)bupt.edu.cn ------------------------------------

5 10

Re: [SR-Users] Question on Database Version Table
by JR Richardson 15 Nov '19

15 Nov '19

Yep version_table Set the name of the table holding the table version. Useful if the proxy is sharing a database within a project and during upgrades. Default value is “version”. Example of usage: version_table="version44" That's what I was looking for, perfect! Yes, I added accountcode and notes to my table for internal use, was just making notes on columns I need on the new version upgrade. Thanks. JR On Tue, Jun 18, 2019 at 2:17 PM Daniel-Constantin Mierla <miconda(a)gmail.com> wrote: > > Hello, > > like Alex said, you can have two version tables, one for each versions > of kamailio you play with and use in config the core parameter to set > the name of that table. > > I am actually writing to say that address table never had accountcode > and notes columns. Likely they were added in your deployment for other > purposes or either the default column names were changed. The default > column names are listed at: > > - > https://kamailio.org/docs/db-tables/kamailio-db-5.2.x.html#gen-db-address > > These were the same for rather long time, iirc the recent version number > changed mainly due to with modifications done to the size, not due to > new columns. > > Cheers, > Daniel > > On 18.06.19 20:50, JR Richardson wrote: > > Hey All, > > > > I'm doing some upgrades and noticed database table version numbers > > have increased. > > > > database table adjustments: > > | address | production ver 3 | upgrade 5.2 ver 6 | need to add > > accountcode and notes column > > | dispatcher | production ver 3 | upgrade 5.2 ver 4 | need to add attr column > > | trusted | production ver 5 | upgrade 5.2 ver 6 | need to add > > ruri_pattern and priority column > > > > I run mysql master/slave so I'm concerned about the master database > > not having the correct version numbers or can I just update the > > version number in the table when I do the upgrade of the kamailio > > versions? > > > > Thanks. > > > > JR

4 6

Re-invite after RTPengine pacemaker switching?
by Жан Базаров 07 Nov '19

07 Nov '19

I need to send re-invite after pacemaker fails over on new rtpengine server. Because new rtpengine dont participate in DTLS handshake and i hear nothing, but silence. I think, may me its would be work. Do you have any idea on this issue?

3 5

kamailio as UAC and NAPTR / sendsocket question
by Karsten Horsmann 09 Aug '19

09 Aug '19

Hi List, I try to register to Deutsche Telekom and there product Deutschland Lan siptrunk. Thats works find but i see an intressting behaviour on selecting the right outgoing interface. Kamailio is sending out with tcp the REQUEST via first private ip configured on that server (172.20.120.53). There is no listen directive for that. I forced NAPTR to use tcp or udp and i assume that kamailio got the right dns answers. On the list i read also that i can use force_send_socket to force the outgoing request. Now my idea - hey i use the $rP for the outgoing to select the right outgoing listen directive. $rP - reference to transport protocol of R-URI But to my surprise the logfile told me thats "UDP" - it sends out via TCP (thats okay). Whats an good transport selector variable from kamailio that works? event_route[tm:local-request] { if(!is_method("OPTIONS")) { xlog("L_INFO", "[tm:local-request] request [$rm] from [$fu] to [$ru] [$rP]\n"); } } INFO: <script>: [tm:local-request] request [REGISTER] from [ sip:+49XXXXXXXX@sip-trunk.telekom.de] to [sip:sip-trunk.telekom.de] [UDP] listen=tcp:2xx.xx.xx.xx:5060 listen=udp:2xx.xx.xx.xx:5060 listen=tls:2xx.xx.xx.xx:5061 advertise CFG_EXT_NAME:5061 listen=udp:172.20.120.55:5060 listen=udp:172.20.120.56:5060 listen=udp:172.20.120.57:5060 listen=udp:172.20.120.58:5060 listen=tcp:172.20.120.58:5060 use_dns_cache=on # use internal DNS cache use_dns_failover=on # depends on internal DNS cache dns_srv_loadbalancing=on dns_try_naptr=on dns_retr_time=1 # seconds before retrying a DNS request dns_retr_no=3 # number of DNS retransmissions dns_naptr_ignore_rfc=yes # ignore target NAPTR priority dns_tcp_pref=30 # TCP has second-highest priority dns_udp_pref=10 # use UDP with least priority tcp_connection_lifetime=3605 # set higher than registration expires #dont' restore modparam("uac","restore_mode","none") modparam("uac","restore_dlg",0) ## UAC REGISTER #!ifdef WITH_UAC_REGISTER modparam("uac", "reg_contact_addr", "CFG_PROD_IP") modparam("uac", "reg_timer_interval", 10) modparam("uac", "reg_retry_interval", 10) modparam("uac", "reg_db_url", DBURL) modparam("uac", "restore_mode", "none") modparam("uac", "auth_username_avp", "$avp(AVP_AUTH_USERNAME)") modparam("uac", "auth_password_avp", "$avp(AVP_AUTH_PASSWORD)") modparam("uac", "auth_realm_avp", "$avp(AVP_AUTH_REALM)") #!endif ip a l 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ens192: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 00:50:56:b5:c1:48 brd ff:ff:ff:ff:ff:ff inet 172.20.120.53/24 brd 172.20.120.255 scope global ens192 valid_lft forever preferred_lft forever inet 2xx.xx.xx.xx/29 scope global ens192 valid_lft forever preferred_lft forever inet 172.20.120.56/24 scope global secondary ens192 valid_lft forever preferred_lft forever inet 172.20.120.57/24 scope global secondary ens192 valid_lft forever preferred_lft forever inet 172.20.120.58/24 scope global secondary ens192 valid_lft forever preferred_lft forever inet 172.20.120.55/24 brd 172.20.120.255 scope global secondary ens192 valid_lft forever preferred_lft forever inet6 fe80::250:56ff:feb5:c148/64 scope link valid_lft forever preferred_lft forever default via 172.20.120.253 dev ens192 172.20.120.0/24 dev ens192 proto kernel scope link src 172.20.120.53 2xx.xx.xx.xx/29 dev ens192 proto kernel scope link src 2xx.xx.xx.xx -- Kind Regards Mit freundlichen Grüßen *Karsten Horsmann*

4 9

100% CPU usage after NTP date update
by Benjamin Marty 07 Aug '19

07 Aug '19

Hello, I'm facing the same issue. Is there a solution for this in the meantime? Benjamin https://lists.kamailio.org/pipermail/sr-users/2015-November/090487.html

2 2

kamailio 4.4.4 fills IPv6 contact address when sending packet over IPv4
by Pali Rohár 24 Jul '19

24 Jul '19

Hi! I detected strange problem that sip.linphone.org refuse to accept presence information re-transmitted by kamailio 4.4.4 server. I debug this problem with tcpdump and I found out that problem is in kamailio which fills IPv6 address into UDP datagram and that datagram is sent via IPv4 socket to IPv4 address, to sip.linphone.org server. And sip.linphone.org server does not have IPv6 connectivity, so correctly return over IPv4 to sender just "400 Bad Contact Header" error. On my server is running kamailio 4.4.4 from Debian Stretch and I can 100% reproduce this problem against public sip.linphone.org server. My server has both IPv4 and IPv6 connectivity and kamailio is listening for both IPv4 and IPv6 connections. So why is kamailio sending IPv6 address over IPv4 and therefore makes it impossible to communicate with non-IPv6 enabled servers? Looks like a problem with choosing default/correct socket for Contact header. And how to fix this problem? Can you help me? I would like to have working interconnection with linphone servers. Just to note I'm seeing this problem only for presence information packets. Other requests, like INVITE or MESSAGE seems to work. Below is relevant tcpdump output. Some parts were replaced by {VAR}. PS: I'm not subscribed to list, so please CC my address when sending reply. Thank you! 17:22:58.121719 IP (tos 0x10, ttl 64, id 21629, offset 0, flags [none], proto UDP (17), length 1266) {MY_IPV4_ADDRESS}.5060 > 91.121.209.194.5060: [bad udp cksum 0xa099 -> 0x9825!] SIP, length: 1238 NOTIFY sip:{REMOTE_NAME}@{REMOTE_USER_IPV4_ADDRESS}:5060;registering_acc=sip_linphone_org SIP/2.0 Via: SIP/2.0/UDP {MY_IPV4_ADDRESS};branch=z9hG4bK2b55.88f93c20000000000000000000000000.0 To: <sip:{REMOTE_NAME}@sip.linphone.org>;tag=75559182 From: <sip:{MY_SIP_URI}>;tag=97d8e785fdf42bf9622a64c13c504961-2708 CSeq: 2 NOTIFY Call-ID: 26cf9d5c019af2dc3302b770887bcc2e@0:0:0:0:0:0:0:0 Route: <sip:91.121.209.194:5060;lr> Content-Length: 597 User-Agent: kamailio (4.4.4 (x86_64/linux)) Max-Forwards: 70 Event: presence Contact: <sip:{MY_IPV6_ADDRESS}:5060;transport=udp> Subscription-State: active;expires=3600 Content-Type: application/pidf+xml <?xml version="1.0" encoding="UTF-8"?> <presence xmlns="urn:ietf:params:xml:ns:pidf" xmlns:dm="urn:ietf:params:xml:ns:pidf:data-model" xmlns:rpid="urn:ietf:params:xml:ns:pidf:rpid" entity="Pali <sip:{MY_SIP_URI}>"> <tuple id="sg89ae"> <status><basic>open</basic></status> <contact priority="0.8">Pali <sip:{MY_SIP_URI}></contact> </tuple> <tuple xmlns="urn:ietf:params:xml:ns:pidf" id="TA0C538B2"> <status> <basic>closed</basic> </status> <contact priority="1">sip:{MY_SIP_URI}</contact> <timestamp>2019-04-19T17:20:36+02:00</timestamp> </tuple> </presence> 17:22:58.151188 IP (tos 0x0, ttl 52, id 22949, offset 0, flags [none], proto UDP (17), length 373) 91.121.209.194.5060 > {MY_IPV4_ADDRESS}.5060: [udp sum ok] SIP, length: 345 SIP/2.0 400 Bad Contact Header Via: SIP/2.0/UDP {MY_IPV4_ADDRESS};branch=z9hG4bK2b55.88f93c20000000000000000000000000.0;rport=5060 From: <sip:{MY_SIP_URI}>;tag=97d8e785fdf42bf9622a64c13c504961-2708 To: <sip:{REMOTE_NAME}@sip.linphone.org>;tag=75559182 Call-ID: 26cf9d5c019af2dc3302b770887bcc2e@0:0:0:0:0:0:0:0 CSeq: 2 NOTIFY Content-Length: 0 -- Pali Rohár pali.rohar(a)gmail.com

4 10

UDP fragmentation issues
by Gerry | x164 15 Jul '19

15 Jul '19

Hi, I have setup Kamailio on Centos 7 as a proxy in front of Asterisk in the same datacenter, using the dispatcher module. I have noted that large INVITEs sent by Kamailio do not arrive at the Asterisk box. It seems the network is dropping fragmented UDP traffic. This does not happen however when I contact the Asterisk box directly from outside the datacenter. Then large INVITES are arriving are Asterisk. So the problem is either in Kamailio or in the datacenter. I have analysed the problem with ngrep and tcpdump/wireshark. In normal mode (udp4_raw=no setting) Kamailio is producing UDP packets above 1500 bytes and leaves all fragmentation to the operating system. Enabling raw udp in Kamailio and lowering the MTU in the server didnt help. I even disabled offloading of UDP fragmentation to the NIC to get to the bottom of this. However nothing helps. Only forcing TCP in the dispatcher works when having large INVITES going to Asterisk. In any case, it appears that there is no guarantee that fragmented UDP arrives at all. As far as I am aware, the recommendation is to use TCP if messages are bigger than 1400 bytes. If so, how can I force Kamailio to send messages over 1400 bytes only in TCP mode? Or is there some other way how to deal with this? Thanks for any hints. Gerry

3 3

DMQ: dmq_t_replicate()
by Lars Olsson 02 Jul '19

02 Jul '19

Hi, I am trying to replicate a REGISTER request between several nodes. When I am using dmq_t_replicate(), it ends the execution of the script. It this correct or I have missed anything? I can not find anything in the documentation of the DMQ module the says that it will end the script execution. Is it recommended to use the DMQ module or should I simply forward the request instead as described in https://medium.com/@tumalevich/kamailio-registration-replication-without-dm… Best Regards, Lars

2 3