sr-users July 2018

sr-users@lists.kamailio.org

84 participants
99 discussions

tcp Free fragment not found
by Diego Nadares 04 Oct '18

04 Oct '18

Hi guys! We have four a scripts that sends dlg.list command to kamailio through xmlrpc. They are not executed simultaneously. Sometimes, we are getting the following errors: fm_search_defrag(): fm_search_defrag(0x7f212c25a000, 5273984); Free fragment not found!2017-05-26T13:54:20.438736-03:00 drwsip02 /usr/local/sbin/kamailio[19833]: ERROR: <core> [mem/f_malloc.c:498]: fm_malloc(): fm_malloc(0x7f212c25a000, 5273984) called from core: tcp_main.c: tcpconn_new(957), module: core; Free fragment not found!2017-05-26T13:54:20.439101-03:00 drwsip02 /usr/local/sbin/kamailio[19833]: ERROR: <core> [tcp_main.c:959]: tcpconn_new(): mem. allocation failure2017-05-26T13:54:20.439344-03:00 drwsip02 /usr/local/sbin/kamailio[19833]: ERROR: <core> [tcp_main.c:3985]: handle_new_connect(): tcpconn_new failed, closing socket I all ready increased this two variables: tcp_conn_wq_max=30000000 tcp_rd_buf_size=5273226 How can I see how much memory xmlrpc is using? Any advice will be appreciated. Thanks in advance! Diego

2 2

Module RTPPROXY param Timeout_Socket not Working
by Hamid Hashmi 24 Sep '18

24 Sep '18

Kamailio didn't start listening on IP mentioned in Module RTPPROXY parameter Timeout_Socket<https://kamailio.org/docs/modules/5.1.x/modules/rtpproxy.html#rtpproxy.p.ti…>. Below is my configuration file. loadmodule "rtpproxy.so" modparam("rtpproxy", "rtpproxy_sock", "udp:127.0.0.1:7722") modparam("rtpproxy", "rtpproxy_tout", 1) modparam("rtpproxy", "rtpproxy_retr", 2) modparam("rtpproxy", "timeout_socket", "xmlrpc:http://127.0.0.1:8000/RPC2") loadmodule "xmlrpc.so" modparam("xmlrpc", "route", "XMLRPC"); modparam("xmlrpc", "mode", 1) modparam("xmlrpc", "url_match", "^/RPC2") when I type a command $ netstat -nlp it only shows SIP listening ports. Kamailio should also start listening on 127.0.0.1:8000. Below are the RTPproxy logs, RTPproxy is also unable to bridge the media DBUG:get_command:GLOBAL: received command "19083_4 Uc120,101 9a1060af660d4202976c419ab5300470 58.65.176.42 4000 1ea06d1eb0a8443697a08e568045598c;1" INFO:rtpp_command_ul_handle:GLOBAL: new session 9a1060af660d4202976c419ab5300470, tag 1ea06d1eb0a8443697a08e568045598c;1 requested, type strong INFO:rtpp_command_ul_handle:9a1060af660d4202976c419ab5300470: new session on a port 35012 created, tag 1ea06d1eb0a8443697a08e568045598c;1 INFO:rtpp_command_ul_handle:9a1060af660d4202976c419ab5300470: pre-filling caller's address with 58.65.176.42:4000 DBUG:rtpc_doreply:GLOBAL: sending reply "35012 51.38.201.3 " DBUG:get_command:GLOBAL: received command "19083_5 Uc120,101 9a1060af660d4202976c419ab5300470 58.65.176.42 4022 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq;1 1ea06d1eb0a8443697a08e568045598c;1 xmlrpc:http://127.0.0.1:8000/RPC2" INFO:rtpp_command_ul_handle:GLOBAL: new session 9a1060af660d4202976c419ab5300470, tag 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq;1 requested, type strong INFO:rtpp_command_ul_handle:9a1060af660d4202976c419ab5300470: new session on a port 35014 created, tag 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq;1 ERR:rtpp_command_ul_handle:9a1060af660d4202976c419ab5300470: invalid socket name 1ea06d1eb0a8443697a08e568045598c;1 INFO:rtpp_command_ul_handle:9a1060af660d4202976c419ab5300470: pre-filling caller's address with 58.65.176.42:4022 DBUG:rtpc_doreply:GLOBAL: sending reply "35014 51.38.201.3 " DBUG:get_command:GLOBAL: received command "19086_4 D 9a1060af660d4202976c419ab5300470 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq 1ea06d1eb0a8443697a08e568045598c" INFO:handle_delete:9a1060af660d4202976c419ab5300470: forcefully deleting session 1 on ports 35012/0 INFO:remove_session:9a1060af660d4202976c419ab5300470: RTP stats: 0 in from callee, 0 in from caller, 0 relayed, 0 dropped INFO:remove_session:9a1060af660d4202976c419ab5300470: RTCP stats: 0 in from callee, 0 in from caller, 0 relayed, 0 dropped INFO:remove_session:9a1060af660d4202976c419ab5300470: session on ports 35012/0 is cleaned up INFO:handle_delete:9a1060af660d4202976c419ab5300470: forcefully deleting session 1 on ports 35014/0 INFO:remove_session:9a1060af660d4202976c419ab5300470: RTP stats: 0 in from callee, 0 in from caller, 0 relayed, 0 dropped INFO:remove_session:9a1060af660d4202976c419ab5300470: RTCP stats: 0 in from callee, 0 in from caller, 0 relayed, 0 dropped INFO:remove_session:9a1060af660d4202976c419ab5300470: session on ports 35014/0 is cleaned up DBUG:rtpc_doreply:GLOBAL: sending reply "0 " DBUG:get_command:GLOBAL: received command "19086_5 D 9a1060af660d4202976c419ab5300470 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq 1ea06d1eb0a8443697a08e568045598c" INFO:handle_command:GLOBAL: delete request failed: session 9a1060af660d4202976c419ab5300470, tags 9RPsEBbcZSA23MmWAw0hfXzJ-bG1GYDq/1ea06d1eb0a8443697a08e568045598c not found DBUG:rtpc_doreply:GLOBAL: sending reply "E50 But if I comment out the param "Timeout_Socket" then there is no issue in bridging the media. Regards Hamid R. Hashmi

3 6

DMQ mem leak issues
by Rogelio Perez 04 Sep '18

04 Sep '18

Hello, We're running three instances of Kamailo v5.14 as registrars handling registrations from ~2000 SIP clients, with one instance being primary and the other two as backups. The three of them are using the dmq and dmq_usrloc modules to synchronize user locations, however after a couple of days of operation the two failover instances show memory leak behaviors, with mem usage assigned to the core taking all available resources. When this happens we've noticed that: - The shared memory used by the function "sip_msg_shm_clone" spikes (from 1kb to 1.5GB). - The shared memory used by the function "dmq:worker.c:job_queue_push" also increases, but not as much (from 1kb to 1MB) - DMQ request are not being answered (with a 200 OK) by the affected instance during this memory leak, which make us think that DMQ module becomes unresponsive. A few more notes: - The failover instances are doing nothing except receiving replicated contacts. - The shared memory grows at the same rate on both instances, but the critical behavior never happens at the same time. - We are allocating 1GB memory on startup to each instance. - We store the location DB in a psql DB and we load it at startup. - We didn't find any errors in syslog, even at debug level. Has anyone experienced a similar issue who can suggest a possible solution? Thanks, Rogelio Perez Telnyx

6 23

MYSQL and various
by Laura 12 Aug '18

12 Aug '18

Dear list, on my kamailio 4.4.7 i use interactions with MYSQL. For example .. i use it for some black list control and other like this.. sql_query("cb", "select id from blacklist where cli='$var(g2)' and (ippeer = '$si' or ippeer = 'ALL')", "rabl"); if($dbr(rabl=>rows)>0){ sql_result_free("rabl"); sl_send_reply("403","Forbidden blacklist"); exit; } sql_result_free("rabl"); I was wonder what was the best method to prevent the query going on if mysql don't respond for high load or other situation.. If Mysql is not able to reply to the query tight now the call continue over the configuration... what is the best way to prevent this and for examples drop it with a standard SIP 503 error. I have search on docs.. but i haven't found any valid way. Thanks Laura

2 1

kamcmd htable.reload propertiesHash problem
by Ginhoux, Patrick 10 Aug '18

10 Aug '18

Hi, I run Kamailio 5.0.1 with MariaDB database. There is no issue at the initialization, meaning the db is opened by kamailio and the hash tables loaded from the db tables. Example: # kamcmd htable.dump propertiesHash { entry: 1 size: 1 slot: { item: { name: cfg.env But the htable.reload command fails: error: 500 - Failed to open htable db connection The Kamailio log reports that it can't connect the db : DEBUG: ctl [../../core/io_wait.h:380]: io_watch_add(): DBG: io_watch_add(0x7f68c74ab300, 10, 3, 0xdc9410), fd_no=1 DEBUG: ctl [io_listener.c:442]: handle_new_connect(): handle_stream read: new connection (1) on /var/run/kamailio//kamailio_ctl DEBUG: <core> [core/sr_module.c:694]: find_mod_export_record(): find_export_record: found <db_bind_api> in module db_mysql [/usr/lib64/kamailio/modules/db_mysql.so] DEBUG: <core> [db.c:209]: db_bind_mod(): using db bind api for db_mysql DEBUG: <core> [db.c:314]: db_do_init2(): connection 0x7f68cc354c10 not found in pool DEBUG: db_mysql [km_my_con.c:98]: db_mysql_new_connection(): opening connection: mysql://xxxx:xxxx@/siprouter ERROR: db_mysql [km_my_con.c:124]: db_mysql_new_connection(): driver error: Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2 "No such file or directory") ERROR: <core> [db.c:318]: db_do_init2(): could not add connection to the pool ERROR: htable [ht_db.c:92]: ht_db_open_con(): failed to connect to the database My local db is located at /myqdata/mysiprdb/data where is the mysiprdb.sock file. But the log shows that Kamailio is looking for /var/lib/mysql/mysql.sock. Does this error make sense to you ? Is it possible to fix this problem by configuration in Kamailio ? My kamailio.cfg script contains : # ----- htable modparam("htable", "db_url", "mysql://[kamailio]/siprouter") modparam("htable", "fetch_rows", 1000) modparam("htable", "htable", "mbxrangeHash=>size=4;dbtable=mbxrange;") modparam("htable", "htable", "abpqHash=>size=12;dbtable=abpq;") modparam("htable", "htable", "zabpqHash=>size=12;dbtable=zabpq;") modparam("htable", "htable", "propertiesHash=>size=4;dbtable=properties;") In the /myqdata/mysiprdb/data/my.cnf, I added the following entries for Kamailio: [kamailio] socket = /myqdata/mysiprdb/data/mysiprdb.sock user = siprouter password = ########## default-character-set = utf8 Thanks all for help. Cordialement. Patrick GINHOUX

2 12

Kamailio troubleshooting
by Sergey Safarov 09 Aug '18

09 Aug '18

One my server with perl external script time to time is stopping processing SIP messages. Perl script make calls to kamailio database on mysql. I have found this in logs DEBUG: <core> [core/io_wait.h:380]: io_watch_add(): DBG: io_watch_add(0x56427eb40c40, 495, 2, 0x7f1a9f9ba408), fd_no=33 DEBUG: <core> [core/io_wait.h:602]: io_watch_del(): DBG: io_watch_del (0x56427eb40c40, 495, -1, 0x0) fd_no=34 called DEBUG: <core> [core/tcp_main.c:4196]: handle_tcpconn_ev(): sending to child, events 1 DEBUG: <core> [core/tcp_main.c:3871]: send2child(): WARNING: no free tcp receiver, connection passed to the least busy one (56) DEBUG: <core> [core/tcp_main.c:3875]: send2child(): selected tcp worker 4 20(24) for activity on [tcp:0.0.0.0:81], 0x7f1a9f9ba408 And no next messages. How get info what is load on "tcp receiver". This host in dev mode and handles not more 5 devices. Sergey

2 2

Kamctl Stats Update
by Noah Mehl 07 Aug '18

07 Aug '18

Hey all, I was wondering if the kamctl stats output could be updated to provide a bit more parseable JSON? For instance: # kamctl stats shmem | jq . { "jsonrpc": "2.0", "result": [ "shmem:fragments = 18", "shmem:free_size = 467187808", "shmem:max_used_size = 69694104", "shmem:real_used_size = 69683104", "shmem:total_size = 536870912", "shmem:used_size = 41048984" ], "id": 4300 } I would hope would be more like this: { "jsonrpc":"2.0", "result":[ { "shmem":[ { "fragments":18, "free_size":467187808, "max_used_size":69694104, "real_used_size":69683104, "total_size":536870912, "used_size":41048984 } ] } ], "id":4300 } I apologize in advanced because I don’t have the skillset to contribute such a feature. Thanks! ~Noah

5 8

DMQ problems
by Aleksandar Sosic 05 Aug '18

05 Aug '18

Hi, I've set up a dmq server for other kamailio nodes with this parameters: ``` modparam("dmq", "server_address","sip:10.0.0.101:5060") modparam("dmq", "notification_address", "sip:127.0.0.1:5060") modparam("dmq", "multi_notify", 1) modparam("dmq", "num_workers", 4) modparam("dmq", "ping_interval", 15) ``` On the kamailio nodes the config is the following: ``` modparam("dmq", "server_address", "sip:10.0.0.x:5060") modparam("dmq", "notification_address", "sip:10.0.0.101:5060") modparam("dmq", "multi_notify", 1) modparam("dmq", "num_workers", 4) modparam("dmq", "ping_interval", 15) ``` On the kamailio dmq server I've got this in the logs: ``` 25(41) ERROR: dmq [notification_peer.c:596]: notification_resp_callback_f(): deleting server sip:127.0.0.1:5060 because of failed request 25(41) ERROR: dmq [notification_peer.c:599]: notification_resp_callback_f(): not deleting notification_peer ``` On the kamailio nodes the same thing with the ip of the dmq server: ``` 25(112) ERROR: dmq [notification_peer.c:596]: notification_resp_callback_f(): deleting server sip:10.0.0.101:5060 because of failed request 25(112) ERROR: dmq [notification_peer.c:599]: notification_resp_callback_f(): not deleting notification_peer ``` The request route on the dmq server is simple as this: ``` request_route { if(is_method("KDMQ")){ dmq_handle_message(); } }#end request_route ``` I've tried with some xlog-s inside the if(is_method("KDMQ")) but they don't show up so the dmq_handle_message() is not even executed. I've also tried to bind the dmq server to a different port but with no success on making it work. The dmq.list_nodes shows only the dmq server IP and the current node IP, the IPs of other nodes are not shown: ``` root@proxy-94b6ccf46-t8fx2:/# kamcmd dmq.list_nodes { host: 10.0.0.101 port: 5060 resolved_ip: 10.0.0.101 status: active last_notification: 0 local: 0 } { host: 10.0.0.45 port: 5060 resolved_ip: 10.0.0.45 status: active last_notification: 0 local: 1 } ``` And it's the same on all other kamailio nodes: dmq server ip and local ip. The kamailio version on all the nodes is: 5.2.0~dev6+0~20180616010152.1138+xenial Thanks, -- Aleksandar Sosic mail: alex.sosic(a)evosip.cloud

4 20

Security announcement related to Kamailio
by Henning Westerholt 04 Aug '18

04 Aug '18

Hello, I want to highlight that the last stable versions (for the two maintained series: 5.0 and 5.1) include fixes for an security issues that can crash a running instance of Kamailio, therefore it is strongly recommended to upgrade. Details: There exists a security vulnerability in the Kamailio SIP server related to "To" header processing. A specially crafted SIP message with double "To" header and an empty "To" tag causes a segmentation fault and crashes Kamailio. The reason is missing input validation in the "build_res_buf_from_sip_req" core function. If an attacker sends many of this messages this would lead to a Denial of Service of the attacked infrastructure. This is especially critical as no authentication for the remote source is needed. You find all the details including a proof of concept code in the published security announcement in my blog: https://skalatan.de/blog/advisory-hw-2018-05 A CVE report for this issue is going to be created as well in the near future. The issues were found some months ago and were fixed quickly. The code related to the reported issues is rather old and there are no known incidents of exploiting these issues so far. However, once the CVE report becomes public, there could be a higher risk of exploitation. Please address any detailed technical questions related to this to the developer list at sr-dev(a)lists.kamailio.org . In case of confidential remarks related to this or other security issues, please address them to the Kamailio Management. Best regards, Henning Westerholt -- Henning Westerholt https://skalatan.de/blog/

1 1

is_method not matching message type KDMQ breaking DMQ bus. (master branch)
by Asgaroth 02 Aug '18

02 Aug '18

Hi All, I've come across and oddity whereby my is_method("KDMQ") checks are not matching. I'm testing with kamailio master: version: kamailio 5.2.0-dev6 (x86_64/linux) a16272 flags: STATS: Off, USE_TCP, USE_TLS, USE_SCTP, TLS_HOOKS, DISABLE_NAGLE, USE_MCAST, DNS_IP_HACK, SHM_MEM, 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_BLACKLIST, 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. id: a16272 compiled on 15:27:21 Jul 31 2018 with gcc 7.3.1 Sample routing logic: request_route { route("InitialMessageChecks"); exit; } route[InitialMessageChecks] { xlog("L_INFO", "Received message type $rm from $si:$sp on $Ri:$Rp"); if ( is_method("KDMQ") ) { xlog("L_INFO", "KDMQ check passed."); } if ( is_method("OPTIONS") ) { xlog("L_INFO", "OPTIONS check passed."); } if ( is_method("SUBSCRIBE") ) { xlog("L_INFO", "SUBSCRIBE check passed."); } } In the logs I am only receiving the 1st log line in the initial message checks route for message types of KDMQ, eg: INFO: <script>: Received message KDMQ from 1.1.1.1:5060 on 1.1.1.2:5060 INFO: <script>: Received message KDMQ from 1.1.1.3:5060 on 1.1.1.2:5060 I never seem to hit the is_method check log line when the method type is KDMQ. I do see the options/subscribe logs though: INFO: <script>: OPTIONS check passed. INFO: <script>: SUBSCRIBE check passed. nothing for KDMQ. Reason for asking is: this check never matches and breaks the dmq bus: if ( is_method("KDMQ") ) { if ( dmq_is_from_node() ) { dmq_handle_message(); exit; } else { exit; } } Any thoughts, or am I going crazy :) Thanks

3 4

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

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 July 2018