sr-dev January 2015

sr-dev@lists.kamailio.org

38 participants
530 discussions

git:master:461aad8c: Add Bandwidth AVP to Rx-Requests as per 3GPP TS 29.213

by Carsten Bock

Module: kamailio Branch: master Commit: 461aad8cae845b8875e961ed3068d9c762d973a8 URL: https://github.com/kamailio/kamailio/commit/461aad8cae845b8875e961ed3068d9c… Author: Carsten Bock <carsten(a)ng-voice.com> Committer: Carsten Bock <carsten(a)ng-voice.com> Date: 2015-01-30T13:03:28+02:00 Add Bandwidth AVP to Rx-Requests as per 3GPP TS 29.213 --- Modified: modules/ims_qos/doc/ims_qos.xml Modified: modules/ims_qos/doc/ims_qos_admin.xml Modified: modules/ims_qos/mod.c Modified: modules/ims_qos/rx_avp.c Modified: modules/ims_qos/rx_avp.h --- Diff: https://github.com/kamailio/kamailio/commit/461aad8cae845b8875e961ed3068d9c… Patch: https://github.com/kamailio/kamailio/commit/461aad8cae845b8875e961ed3068d9c…

9 years, 10 months

git:master:85d3488c: sl: re-attempt to bind to tm after all mods init

by Daniel-Constantin Mierla

Module: kamailio Branch: master Commit: 85d3488c3910d447f6900df166bc6b0c16a64e9c URL: https://github.com/kamailio/kamailio/commit/85d3488c3910d447f6900df166bc6b0… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: 2015-01-30T10:30:16+01:00 sl: re-attempt to bind to tm after all mods init - handle the situation when tm is initialized after the sl module - re-binding is done on child callback with rank PROC_INIT --- Modified: modules/sl/sl.c --- Diff: https://github.com/kamailio/kamailio/commit/85d3488c3910d447f6900df166bc6b0… Patch: https://github.com/kamailio/kamailio/commit/85d3488c3910d447f6900df166bc6b0… --- diff --git a/modules/sl/sl.c b/modules/sl/sl.c index 5f3febb..27c30e8 100644 --- a/modules/sl/sl.c +++ b/modules/sl/sl.c @@ -164,13 +164,13 @@ static int mod_init(void) return -1; } + memset(&tmb, 0, sizeof(struct tm_binds)); if(sl_bind_tm!=0) { if(load_tm_api(&tmb)==-1) { LM_INFO("could not bind tm module - only stateless mode" - " available\n"); - sl_bind_tm=0; + " available during modules initialization\n"); } } @@ -186,6 +186,14 @@ static int child_init(int rank) ERR("init_sl_stats_child failed\n"); return -1; } + if(sl_bind_tm!=0 && tmb.register_tmcb==0) { + if(load_tm_api(&tmb)==-1) { + LM_INFO("could not bind tm module - only stateless mode" + " available during runtime\n"); + sl_bind_tm=0; + } + } + } return 0; } @@ -267,7 +275,7 @@ int send_reply(struct sip_msg *msg, int code, str *reason) } } - if(sl_bind_tm!=0) + if(sl_bind_tm!=0 && tmb.t_gett!=0) { t = tmb.t_gett(); if(t!= NULL && t!=T_UNDEFINED) @@ -324,7 +332,7 @@ int get_reply_totag(struct sip_msg *msg, str *totag) struct cell * t; if(msg==NULL || totag==NULL) return -1; - if(sl_bind_tm!=0) + if(sl_bind_tm!=0 && tmb.t_gett!=0) { t = tmb.t_gett(); if(t!= NULL && t!=T_UNDEFINED)

9 years, 10 months

git:master:1d3b89b1: db_mysql: restore group ending character in db url

by Daniel-Constantin Mierla

Module: kamailio Branch: master Commit: 1d3b89b10bfd825e4ca73bb95b0ae489843dd9f1 URL: https://github.com/kamailio/kamailio/commit/1d3b89b10bfd825e4ca73bb95b0ae48… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: 2015-01-30T10:30:16+01:00 db_mysql: restore group ending character in db url - ] was set to \0 to use it for mysql options, but it will break matching the db url for connection pooling - reported by Juha Heinanen --- Modified: modules/db_mysql/km_my_con.c --- Diff: https://github.com/kamailio/kamailio/commit/1d3b89b10bfd825e4ca73bb95b0ae48… Patch: https://github.com/kamailio/kamailio/commit/1d3b89b10bfd825e4ca73bb95b0ae48… --- diff --git a/modules/db_mysql/km_my_con.c b/modules/db_mysql/km_my_con.c index d0338a4..deadd76 100644 --- a/modules/db_mysql/km_my_con.c +++ b/modules/db_mysql/km_my_con.c @@ -43,7 +43,7 @@ struct my_con* db_mysql_new_connection(const struct db_id* id) { struct my_con* ptr; - char *host, *grp; + char *host, *grp, *egrp; unsigned int connection_flag = 0; if (!id) { @@ -57,6 +57,7 @@ struct my_con* db_mysql_new_connection(const struct db_id* id) return 0; } + egrp = 0; memset(ptr, 0, sizeof(struct my_con)); ptr->ref = 1; @@ -68,9 +69,10 @@ struct my_con* db_mysql_new_connection(const struct db_id* id) mysql_init(ptr->con); - if (id->host[0] == '[' && (host = strchr(id->host, ']')) != NULL) { + if (id->host[0] == '[' && (egrp = strchr(id->host, ']')) != NULL) { grp = id->host + 1; - *host = '\0'; + *egrp = '\0'; + host = egrp; if (host != id->host + strlen(id->host)-1) { host += 1; // host found after closing bracket } @@ -129,11 +131,13 @@ struct my_con* db_mysql_new_connection(const struct db_id* id) ptr->timestamp = time(0); ptr->id = (struct db_id*)id; + if(egrp) *egrp = ']'; return ptr; err: if (ptr && ptr->con) pkg_free(ptr->con); if (ptr) pkg_free(ptr); + if(egrp) *egrp = ']'; return 0; }

9 years, 10 months

git:4.2:e8a30618: presence_dialoginfo : optionally send dummy dialog

by Luis Azedo

Module: kamailio Branch: 4.2 Commit: e8a306188cf50efb7b4a352d65494b4c736c9584 URL: https://github.com/kamailio/kamailio/commit/e8a306188cf50efb7b4a352d65494b4… Author: lazedo <luis.azedo(a)factorlusitano.com> Committer: Luis Azedo <luis(a)2600hz.com> Date: 2015-01-30T05:39:10Z presence_dialoginfo : optionally send dummy dialog (cherry picked from commit 5c9b3afca092dfda33c96dbf5195b8f7bbc2d4cf) --- Modified: modules/presence_dialoginfo/doc/presence_dialoginfo_admin.xml Modified: modules/presence_dialoginfo/notify_body.c Modified: modules/presence_dialoginfo/presence_dialoginfo.c --- Diff: https://github.com/kamailio/kamailio/commit/e8a306188cf50efb7b4a352d65494b4… Patch: https://github.com/kamailio/kamailio/commit/e8a306188cf50efb7b4a352d65494b4…

9 years, 10 months

request for backporting to 4.2 - presence_dialoginfo : optionally send dummy dialog

by Luis Azedo

Hi List, some sip clients like Bria and some sip phones like Astra expect a <dialog-info><dialog></dialog></dialog-info> on the NOTIFY message sent. presence sends a NOTIFY after the SUBSCRIBE with NULL body if it doesn't find the presentity. the patch already in master allows to configure the creation of a dummy dialog to be sent when presence doesn't find the presentity. this behavior was only observed in dialog subscriptions so the patch was made against presence_dialoginfo. the new param force_dummy_dialog defaults to 0 to preserve existing behavior. using this patch Bria, Astra started working with BLF and others (Cisco, Polycom, Jitsi) kept running. does anyone have anything against backporting this to 4.2 ? Best

9 years, 10 months

git:master:83caaabd: rtpengine: porting rtp_inst_pvar from rtpproxy

by lucian balanceanu

Module: kamailio Branch: master Commit: 83caaabd4725b46d60577280ced436bb9638c80f URL: https://github.com/kamailio/kamailio/commit/83caaabd4725b46d60577280ced436b… Author: lucian balanceanu <lucian.balanceanu(a)1and1.ro> Committer: lucian balanceanu <lucian.balanceanu(a)1and1.ro> Date: 2015-01-29T19:23:26+02:00 rtpengine: porting rtp_inst_pvar from rtpproxy - rtp_inst_pvar: pseudo variable to store the chosen RTP engine address --- Modified: modules/rtpengine/doc/rtpengine_admin.xml Modified: modules/rtpengine/rtpengine.c Modified: modules/rtpengine/rtpengine.h Modified: modules/rtpengine/rtpengine_db.c --- Diff: https://github.com/kamailio/kamailio/commit/83caaabd4725b46d60577280ced436b… Patch: https://github.com/kamailio/kamailio/commit/83caaabd4725b46d60577280ced436b…

9 years, 10 months

git:master:df2a4554: sca: more error handling during initialization of the module

by Daniel-Constantin Mierla

Module: kamailio Branch: master Commit: df2a45541c5e37907374c5e7595de1f30ea84f98 URL: https://github.com/kamailio/kamailio/commit/df2a45541c5e37907374c5e7595de1f… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: 2015-01-29T10:45:52+01:00 sca: more error handling during initialization of the module - fix crash when the module doesn't initialize and calls destroy function during shutdonw at startup - reported by Olle E. Johansson, GH#57 --- Modified: modules/sca/sca.c --- Diff: https://github.com/kamailio/kamailio/commit/df2a45541c5e37907374c5e7595de1f… Patch: https://github.com/kamailio/kamailio/commit/df2a45541c5e37907374c5e7595de1f… --- diff --git a/modules/sca/sca.c b/modules/sca/sca.c index e739e18..920024b 100644 --- a/modules/sca/sca.c +++ b/modules/sca/sca.c @@ -210,6 +210,7 @@ sca_set_config( sca_mod *scam ) LM_ERR( "Failed to shm_malloc module configuration" ); return( -1 ); } + memset(scam->cfg, 0, sizeof( sca_config )); if ( outbound_proxy.s ) { scam->cfg->outbound_proxy = &outbound_proxy; @@ -293,23 +294,23 @@ sca_mod_init( void ) if ( rpc_register_array( sca_rpc ) != 0 ) { LM_ERR( "Failed to register RPC commands" ); - return( -1 ); + goto error; } if ( sca_bind_srdb1( sca, &dbf ) != 0 ) { LM_ERR( "Failed to initialize required DB API" ); - return( -1 ); + goto error; } if ( load_tm_api( &tmb ) != 0 ) { LM_ERR( "Failed to initialize required tm API" ); - return( -1 ); + goto error; } sca->tm_api = &tmb; if ( sca_bind_sl( sca, &slb ) != 0 ) { LM_ERR( "Failed to initialize required sl API" ); - return( -1 ); + goto error; } if ( sca_hash_table_create( &sca->subscriptions,

9 years, 10 months

git:master:8c3a12f6: dialog: proper handling cseq increment when cseq header is before via

by Daniel-Constantin Mierla

Module: kamailio Branch: master Commit: 8c3a12f6aca7c6492ee440da469834997ea89eee URL: https://github.com/kamailio/kamailio/commit/8c3a12f6aca7c6492ee440da4698349… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: 2015-01-29T10:37:13+01:00 dialog: proper handling cseq increment when cseq header is before via - reported by Fernando Aquilino --- Modified: modules/dialog/dlg_cseq.c --- Diff: https://github.com/kamailio/kamailio/commit/8c3a12f6aca7c6492ee440da4698349… Patch: https://github.com/kamailio/kamailio/commit/8c3a12f6aca7c6492ee440da4698349… --- diff --git a/modules/dialog/dlg_cseq.c b/modules/dialog/dlg_cseq.c index 0d39aac..58a6bb8 100644 --- a/modules/dialog/dlg_cseq.c +++ b/modules/dialog/dlg_cseq.c @@ -285,31 +285,60 @@ int dlg_cseq_msg_sent(void *data) /* new cseq value */ dlg->iflags |= DLG_IFLAG_CSEQ_DIFF; - /* copy first part till after via branch */ - tbuf_len = via->branch->value.s + via->branch->value.len - msg.buf; - memcpy(tbuf, msg.buf, tbuf_len); - /* complete via branch */ - tbuf[tbuf_len++] = '.'; - tbuf[tbuf_len++] = 'c'; - tbuf[tbuf_len++] = 's'; - memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s, get_cseq(&msg)->number.len); - tbuf_len += get_cseq(&msg)->number.len; - /* copy till beginning of cseq number */ - memcpy(tbuf+tbuf_len, via->branch->value.s + via->branch->value.len, - get_cseq(&msg)->number.s - via->branch->value.s - - via->branch->value.len); - tbuf_len += get_cseq(&msg)->number.s - via->branch->value.s + + if(via->branch->value.s<get_cseq(&msg)->number.s) { + /* Via is before CSeq */ + /* copy first part till after via branch */ + tbuf_len = via->branch->value.s + via->branch->value.len - msg.buf; + memcpy(tbuf, msg.buf, tbuf_len); + /* complete via branch */ + tbuf[tbuf_len++] = '.'; + tbuf[tbuf_len++] = 'c'; + tbuf[tbuf_len++] = 's'; + memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s, get_cseq(&msg)->number.len); + tbuf_len += get_cseq(&msg)->number.len; + /* copy till beginning of cseq number */ + memcpy(tbuf+tbuf_len, via->branch->value.s + via->branch->value.len, + get_cseq(&msg)->number.s - via->branch->value.s + - via->branch->value.len); + tbuf_len += get_cseq(&msg)->number.s - via->branch->value.s - via->branch->value.len; - /* add new value */ - memcpy(tbuf+tbuf_len, nval.s, nval.len); - tbuf_len += nval.len; - /* copy from after cseq number to the end of sip message */ - memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s+get_cseq(&msg)->number.len, - msg.buf + msg.len - get_cseq(&msg)->number.s - - get_cseq(&msg)->number.len); - tbuf_len += msg.buf+msg.len - get_cseq(&msg)->number.s + /* add new value */ + memcpy(tbuf+tbuf_len, nval.s, nval.len); + tbuf_len += nval.len; + /* copy from after cseq number to the end of sip message */ + memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s+get_cseq(&msg)->number.len, + msg.buf + msg.len - get_cseq(&msg)->number.s + - get_cseq(&msg)->number.len); + tbuf_len += msg.buf+msg.len - get_cseq(&msg)->number.s - get_cseq(&msg)->number.len; - + } else { + /* CSeq is before Via */ + /* copy till beginning of cseq number */ + tbuf_len = get_cseq(&msg)->number.s - msg.buf; + memcpy(tbuf, msg.buf, tbuf_len); + /* add new value */ + memcpy(tbuf+tbuf_len, nval.s, nval.len); + tbuf_len += nval.len; + /* copy from after cseq number to the after via branch */ + memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s+get_cseq(&msg)->number.len, + via->branch->value.s + via->branch->value.len + - get_cseq(&msg)->number.s - get_cseq(&msg)->number.len); + tbuf_len += via->branch->value.s + via->branch->value.len + - get_cseq(&msg)->number.s - get_cseq(&msg)->number.len; + /* complete via branch */ + tbuf[tbuf_len++] = '.'; + tbuf[tbuf_len++] = 'c'; + tbuf[tbuf_len++] = 's'; + memcpy(tbuf+tbuf_len, get_cseq(&msg)->number.s, get_cseq(&msg)->number.len); + tbuf_len += get_cseq(&msg)->number.len; + /* copy from after via to the end of sip message */ + memcpy(tbuf+tbuf_len, via->branch->value.s + via->branch->value.len, + msg.buf + msg.len - via->branch->value.s + - via->branch->value.len); + tbuf_len += msg.buf+msg.len - via->branch->value.s + - via->branch->value.len; + } /* replace old msg content */ obuf->s = pkg_malloc((tbuf_len+1)*sizeof(char)); if(obuf->s==NULL) {

9 years, 10 months

[kamailio] sca: Core dump (#35)

by Olle E. Johansson

If I load the sca module in a minimal configuration without any sca modparams, it core dumps. Thread 0 Crashed:: Dispatch queue: com.apple.main-thread 0 sca.so 0x000000010de29823 sca_db_get_connection + 35 (sca_db.c:158) 1 sca.so 0x000000010de411e7 sca_subscription_db_update + 55 (sca_subscribe.c:493) 2 sca.so 0x000000010ddd44e3 sca_mod_destroy + 19 (sca.c:367) 3 kamailio 0x000000010d1ab089 destroy_modules + 73 (sr_module.c:790) --- Reply to this email directly or view it on GitHub: https://github.com/kamailio/kamailio/issues/35

9 years, 10 months

strict routing and sip spirals

by Jasmin Schnatterbeck

Hi, may there is a routing-issue in rr-module, that causes some problems with sip-spirals: If is_myself("$ru") is true (in-dialog), kamailio rewrites the $ru to the $route_uri (strict routing). That is usually correct. loose.c 944 if (is_myself(&_m->parsed_uri)) { 945 return after_strict(_m); This includes the case if is_myself("$route_uri") is also true - but(!) $route_uri and $ru can still be different, and this can be very relevant. rfc3261 says: 16.4 Route Information Preprocessing The proxy MUST inspect the Request-URI of the request. If the Request-URI of the request contains a value this proxy previously placed into a Record-Route header field (see Section 16.6 item 4), the proxy MUST replace the Request-URI in the request with the last value from the Route header field, and remove that value from the Route header field. The proxy MUST then proceed as if it received this modified request. It says " If the Request-URI of the request contains a value this proxy previously placed into a Record-Route header field ". This means $route_uri == $ru Then - indeed - it does not matter to replace the $ru (even: does it really change something?) I do not see a sign to decide for strict routing instead of loose routing, if (in-dialog) is_myself("$ru") AND is_myself("$route_uri") in general, and especially not if $ru == $route_uri Or is there any reason to process the request like it is now? Cheers Jasmin

9 years, 10 months

← Newer
1
2
3
4
5
6
7
8
...
53
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

sr-dev January 2015