sr-dev December 2013

sr-dev@lists.kamailio.org

30 participants
318 discussions

Outbound flow failure handling merged to master

by Hugh Waite

Hello, As you will see I have merged my branch back into master. These changes add a new event route [tm:branch-failure] to the tm module which is run when any failure response is received on a transaction. The event_route uses a new route type BRANCH_ROUTE which limits the functions that can be run in the route. The functions t_check_status(), t_next_contact_flow(), t_relay() and unregister() can be used in this route. A new pv $T_reply_ruid is accessible in this route which can be used to unregister a single contact entry. The following example route can be used on a registrar to handle failed or invalid flows: event_route[tm:branch-failure] { xlog("L_INFO", "event_route[tm:branch-failure]\n"); if (t_check_status("430|403")) { if (!unregister("location", "$tu", "$T_reply_ruid")) { xlog("L_WARN", "failed to unregister $tu with ruid $T_reply_ruid\n"); } if (!t_next_contact_flow()) { xlog("L_INFO", "No more flows\n"); } else { xlog("L_INFO", "Next flow\n"); t_relay(); } } } Any bugs, memory leaks etc. let me know! Hugh -- Hugh Waite Principal Design Engineer Crocodile RCS Ltd.

6 years, 4 months

Fw: query related to NOTIFY message size

by Piyush.Bansal＠relianceada.com

please help ----- Forwarded by Piyush Bansal/RCOM/RelianceADA on 10/10/2013 10:51 AM ----- From: Piyush Bansal/RCOM/RelianceADA To: serusers(a)iptel.org, serdev(a)iptel.org Date: 10/09/2013 12:14 PM Subject: query related to NOTIFY message size Hi there, I have certain queries regarding batch SUBSCRIBE and NOTIFY. I have a user who has 100 buddies in his buddy list. If any of his buddy changes his/her presence status, that user gets a NOTIFY message with presence status of all the 100 buddies. In that case, the message size is exceeding 500 KB. Thats quite a higher value for a UDP packet. Can anybody suggest- 1. If there is any way to restrict the size of the packet. 2. How to ensure that the packet is received correctly by the client. Thanks and Regards, --Piyush Bansal The information contained in this electronic message (email) and any attachments to this email are intended for the exclusive use of the addressee(s) and access to this email by any one else is unauthorised. The email may contain proprietary, confidential or privileged information or information relating to Reliance Group. If you are not the intended recipient, please notify the sender by telephone, fax, or return email and delete this communication and any attachments thereto, immediately from your computer. Any dissemination, distribution, or copying of this communication and the attachments thereto (in whole or part), in any manner, is strictly prohibited and actionable at law. The recipient acknowledges that emails are susceptible to alteration and their integrity can not be guaranteed and that Company does not guarantee that any e-mail is virus-free and accept no liability for any damage caused by any virus transmitted by this email.

8 years, 5 months

Kamailio on OS/X Mavericks

by Olle E. Johansson

Hi! Kamailio by default doesn't compile on OS/X Mavericks. $ make generating autover.h ... /Applications/Xcode.app/Contents/Developer/usr/bin/make --no-print-directory -wC . cfg-defs target architecture <x86_64>, host architecture <x86_64> making config... CC (gcc) [kamailio] action.o error: invalid value '9' in '-O9' make: *** [action.o] Error 1 $ gcc --version Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.9.sdk/usr/include/c++/4.2.1 Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn) Target: x86_64-apple-darwin13.0.0 Thread model: posix Anyone that has found an optimal way to patch the Makefile? /O

8 years, 7 months

[PATCH 1/1] websocket: remove libunistring dependency

by Timo Teräs

libunistring is old, slow, messy code and uncompilable on new systems. remove the sole user of it, and replace it with inline utf8 decoder implementation from http://bjoern.hoehrmann.de/utf-8/decoder/dfa/. improves performance and portability as libunistring is not needed. --- @Peter, Hugh: Would it be ok for me to push this? modules/websocket/Makefile | 2 +- modules/websocket/README | 1 - modules/websocket/doc/websocket_admin.xml | 3 -- modules/websocket/utf8_decode.h | 52 +++++++++++++++++++++++++++++++ modules/websocket/ws_frame.c | 4 +-- 5 files changed, 55 insertions(+), 7 deletions(-) create mode 100644 modules/websocket/utf8_decode.h diff --git a/modules/websocket/Makefile b/modules/websocket/Makefile index bb7c809..c686a82 100644 --- a/modules/websocket/Makefile +++ b/modules/websocket/Makefile @@ -27,7 +27,7 @@ else # E.g.: make TLS_HOOKS=1 TLS_EXTRA_LIBS="-lz -lkrb5" endif -LIBS+= $(TLS_EXTRA_LIBS) -lunistring +LIBS+= $(TLS_EXTRA_LIBS) # Static linking, if you'd like to use TLS and WEBSOCKET at the same time # diff --git a/modules/websocket/README b/modules/websocket/README index 49d8693..bdba3e4 100644 --- a/modules/websocket/README +++ b/modules/websocket/README @@ -316,7 +316,6 @@ onreply_route[WS_REPLY] { The following libraries must be installed before running Kamailio with this module loaded: * OpenSSL. - * GNU libunistring. 4. Parameters diff --git a/modules/websocket/doc/websocket_admin.xml b/modules/websocket/doc/websocket_admin.xml index fa7d300..e40dc09 100644 --- a/modules/websocket/doc/websocket_admin.xml +++ b/modules/websocket/doc/websocket_admin.xml @@ -262,9 +262,6 @@ onreply_route[WS_REPLY] { <listitem> <para><emphasis>OpenSSL</emphasis>.</para> </listitem> - <listitem> - <para><emphasis>GNU libunistring</emphasis>.</para> - </listitem> </itemizedlist> </para> </section> diff --git a/modules/websocket/utf8_decode.h b/modules/websocket/utf8_decode.h new file mode 100644 index 0000000..b274fe7 --- /dev/null +++ b/modules/websocket/utf8_decode.h @@ -0,0 +1,52 @@ +#include <stdint.h> +#include <stddef.h> + +// Copyright (c) 2008-2010 Bjoern Hoehrmann <bjoern(a)hoehrmann.de> +// See http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ for details. + +#define UTF8_ACCEPT 0 +#define UTF8_REJECT 12 + +static const uint8_t utf8d[] = { + // The first part of the table maps bytes to character classes that + // to reduce the size of the transition table and create bitmasks. + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, + 8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, + 10,3,3,3,3,3,3,3,3,3,3,3,3,4,3,3, 11,6,6,6,5,8,8,8,8,8,8,8,8,8,8,8, + + // The second part is a transition table that maps a combination + // of a state of the automaton and a character class to a state. + 0,12,24,36,60,96,84,12,12,12,48,72, 12,12,12,12,12,12,12,12,12,12,12,12, + 12, 0,12,12,12,12,12, 0,12, 0,12,12, 12,24,12,12,12,12,12,24,12,24,12,12, + 12,12,12,12,12,12,12,24,12,12,12,12, 12,24,12,12,12,12,12,12,12,24,12,12, + 12,12,12,12,12,12,12,36,12,36,12,12, 12,36,12,12,12,12,12,36,12,36,12,12, + 12,36,12,12,12,12,12,12,12,12,12,12, +}; + +static inline uint32_t decode(uint32_t* state, uint32_t* codep, uint32_t byte) +{ + uint32_t type = utf8d[byte]; + + *codep = (*state != UTF8_ACCEPT) ? + (byte & 0x3fu) | (*codep << 6) : + (0xff >> type) & (byte); + + *state = utf8d[256 + *state + type]; + return *state; +} + +static inline int IsUTF8(uint8_t* s, size_t len) +{ + uint32_t codepoint, state = 0; + + while (len--) + decode(&state, &codepoint, *s++); + + return state == UTF8_ACCEPT; +} + diff --git a/modules/websocket/ws_frame.c b/modules/websocket/ws_frame.c index a3a4cef..3562437 100644 --- a/modules/websocket/ws_frame.c +++ b/modules/websocket/ws_frame.c @@ -22,7 +22,7 @@ */ #include <limits.h> -#include <unistr.h> +#include "utf8_decode.h" #include "../../events.h" #include "../../receive.h" #include "../../stats.h" @@ -695,7 +695,7 @@ int ws_frame_transmit(void *data) frame.fin = 1; /* Can't be sure whether this message is UTF-8 or not so check to see if it "might" be UTF-8 and send as binary if it definitely isn't */ - frame.opcode = (u8_check((uint8_t *) wsev->buf, wsev->len) == NULL) ? + frame.opcode = IsUTF8((uint8_t *) wsev->buf, wsev->len) ? OPCODE_TEXT_FRAME : OPCODE_BINARY_FRAME; frame.payload_len = wsev->len; frame.payload_data = wsev->buf; -- 1.8.5.1

8 years, 8 months

Using t_suspend()/t_continue() multiple times on the same transaction

by Peter Dunkley

Hi, I am trying to use t_suspend()/t_continue() multiple times on the same transaction. Calling t_suspend() more than once works, but the second time I call t_continue() the transaction is killed and a 500 response is sent. It is the "if (branch == t->nr_of_outgoings)" check from the code fragment below (from t_suspend.c:t_continue()) that results in the transaction being killed - you can see the debug/error line I added to determine this in the fragment. Is using t_suspend()/t_continue() multiple times something that should work? Thanks, Peter if (t->uas.status < 200) { /* No final reply has been sent yet. * Check whether or not there is any pending branch. */ for ( branch = 0; branch < t->nr_of_outgoings; branch++ ) { if ((t->uac[branch].request.buffer != NULL) && (t->uac[branch].last_received < 200) ) break; } if (branch == t->nr_of_outgoings) { /* There is not any open branch so there is * no chance that a final response will be received. */ ret = 0; LM_ERR("branch == t->nr_of_outgoings\n"); goto kill_trans; } } -- Peter Dunkley Technical Director Crocodile RCS Ltd

8 years, 8 months

[PATCH] modules/sca: use Call-Info for canonical AoR

by Timo Teräs

RFC3261 8.1.1.2 states that To header "may or may not be the ultimate recipient of the request", and 8.1.1.3 states that From header is "possibly the user's address-of-record". So either of them cannot be fully trusted. This was already noted and fixed partially in commit 1ef4587612806a94 ("modules/sca: reconcile Contact and From URIs in ACK callback"). But similar issues happen when using ENUM and/or calling using local alias (numbers only, or local forward) and the destination is in another SIP domain. To-header contains the initial domain, and not the AoR domain. When handling response to such INVITE a sane way to determine correct AoR is to inspect the Call-Info header's Application Server URI. Thus this changes AoR canonicalization to happens as follow: 1. User portion is taken from Contact header if present, otherwise from the applicable From / To header. 2. Domain/port part is taken from Call-Info header if present, and otherwise from the applicable From / To header. --- This has not fully tested, but this seems to be the root case why SCA does not on certain inter-domain calls that I get. I'd like to get review comments, and possible other fix ideas. I'll let you know how the testing goes, and if looks OK to you can push this to master and relevant stable branches. modules/sca/sca_util.c | 51 +++++++++++++++++++++++++++----------------------- 1 file changed, 28 insertions(+), 23 deletions(-) diff --git a/modules/sca/sca_util.c b/modules/sca/sca_util.c index 143ac47..e209cc0 100644 --- a/modules/sca/sca_util.c +++ b/modules/sca/sca_util.c @@ -26,6 +26,7 @@ #include <assert.h> #include "sca_util.h" +#include "sca_call_info.h" #include "../../parser/sdp/sdp.h" @@ -343,9 +344,13 @@ sca_aor_create_from_info( str *aor, uri_type type, str *user, str *domain, sca_create_canonical_aor_for_ua( sip_msg_t *msg, str *c_aor, int ua_opts ) { struct to_body *tf = NULL; - sip_uri_t c_uri; - str tf_aor = STR_NULL; + str user_portion; + str domain_portion; + str port_portion = STR_NULL; str contact_uri = STR_NULL; + sip_uri_t c_uri; + sca_call_info call_info; + hdr_field_t *call_info_hdr; int rc = -1; assert( msg != NULL ); @@ -373,12 +378,6 @@ sca_create_canonical_aor_for_ua( sip_msg_t *msg, str *c_aor, int ua_opts ) } } - if ( sca_uri_extract_aor( &tf->uri, &tf_aor ) < 0 ) { - LM_ERR( "sca_create_canonical_aor: failed to extract AoR from " - "URI <%.*s>", STR_FMT( &tf->uri )); - goto done; - } - memset( &c_uri, 0, sizeof( sip_uri_t )); if (( rc = sca_get_msg_contact_uri( msg, &contact_uri )) < 0 ) { LM_ERR( "sca_create_canonical_aor: failed to get contact URI from " @@ -394,22 +393,28 @@ sca_create_canonical_aor_for_ua( sip_msg_t *msg, str *c_aor, int ua_opts ) } } - if ( SCA_STR_EMPTY( &c_uri.user ) || - SCA_STR_EQ( &c_uri.user, &tf->parsed_uri.user )) { - /* empty contact header or Contact user matches To/From AoR */ - c_aor->s = (char *)pkg_malloc( tf_aor.len ); - c_aor->len = tf_aor.len; - memcpy( c_aor->s, tf_aor.s, tf_aor.len ); + /* Prefer Contact header user, fallback to To/From */ + if ( SCA_STR_EMPTY( &c_uri.user ) ) + user_portion = tf->parsed_uri.user; + else + user_portion = c_uri.user; + + memset( &call_info, 0, sizeof( sca_call_info )); + call_info_hdr = sca_call_info_header_find( msg->headers ); + if ( !SCA_HEADER_EMPTY( call_info_hdr ) && + sca_call_info_body_parse( &call_info_hdr->body, &call_info ) >= 0 ) { + /* Call-Info present, use the server name as AoR domain */ + domain_portion = call_info.sca_uri; } else { - /* Contact user and To/From user mismatch */ - if ( sca_aor_create_from_info( c_aor, c_uri.type, - &c_uri.user, &tf->parsed_uri.host, - &tf->parsed_uri.port ) < 0 ) { - LM_ERR( "sca_create_canonical_aor: failed to create AoR from " - "Contact <%.*s> and URI <%.*s>", - STR_FMT( &contact_uri ), STR_FMT( &tf_aor )); - goto done; - } + /* Use To/From domain */ + domain_portion = tf->parsed_uri.host; + port_portion = tf->parsed_uri.port; + } + + if ( sca_aor_create_from_info( c_aor, c_uri.type, &user_portion, &domain_portion, &port_portion ) < 0 ) { + LM_ERR( "sca_create_canonical_aor: failed to create canonical AoR " + "user: <%.*s>, domain: <%.*s>", STR_FMT( &user_portion ), STR_FMT( &domain_portion )); + goto done; } rc = 1; -- 1.8.4

8 years, 11 months

Removing carrierroute dependency on libconfuse

by Marius Zbihlei

Hi all, This has been in the back of my head from some time now. Libconfuse is a small project not distributed with major distros and unmaintained for 3.5 years now. As it is only used for reading of a simple cfg file, I want to write a simple parser that does the job(I do not want external deps ... see below why) Benefits: we can now pack carrierroute with the common used modules (as it will not have any external dependencies) and also will remove a lot of problems seen on the list regarding linkage of libconfuse. Any thoughts? Marius

9 years, 1 month

default config: do loose_route() for in-dialog NOTIFY as per new RFC 6665

by Iñaki Baz Castillo

Hi, the new RFC 6665 mandates that the subscription dialog is made *after* the first NOTIFY from the UAS (rather than after the 200 to the initial SUBSCRIBE). This means that the UAC MUST take the route set from the *first* received NOTIFY (instead of taking it from the 200 to the initial SUBSCRIBE). Therefore a proxy compliant with RFC 6665 MUST always add RR to in-dialog NOTIFY requests and thus, I propose that the default SR script file implements it. Regards. -- Iñaki Baz Castillo <ibc(a)aliax.net>

9 years, 1 month

Get SIP Digest working

by Bazler, Marek

Hi, 3GPP's "SIP Digest" is not working in the ims_auth module and stopping of the module before the auth_vector was initialized causing "Segmentation fault". code source : origin/4.1 Regards Marek. This e-mail and any attachment is for authorised use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by, any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. Thank you for understanding.

9 years, 2 months

Various fixes for Kamailio's IMS code

by Rob Day

Hi, I've been trying to set up Kamailio 4.0.5 as a P-CSCF, I-CSCF and S-CSCF today, and hit a few issues which I've fixed. http://sip-router.org/contribute/ suggests that patches should be submitted to this list, so here I am. The first issue I hit was the sem_post issue that has been reported from time to time (http://lists.kamailio.org/pipermail/sr-users/2013-January/076554.html) - it looks as though sem_post is only included in -lpthread on Ubuntu 12.04, not -lrt (https://bugs.launchpad.net/ubuntu/+source/manpages/+bug/874418), so I've updated some Makefiles to reflect that. I subsequently hit a segfault when calling pcscf_save: I've tracked this down to a while loop that was missing curly braces, and so tried to dereference a null pointer. I think there's another bug in this piece of code - because h is NULL at the end of the first while loop, we never go round the second loop, and so never call free_rr. It looks like even if we did, h->parsed has been set to 0 before we try and free it, so nothing would happen. I wasn't sure how the memory management here works, though, and so haven't tried to change this. I also found that the ims_auth module didn't handle standard SIP Digest authentication on the MAR interface, only AKA and Digest-MD5 (which seems to be specific to OpenIMSCore HSS, and not mentioned in the IMS specs). Most of the code to allow SIP Digest authentication was already there, so I just hooked the last piece in to get it to work. I'm not sure whether this counts as a new feature or a bugfix - it's not clear whether this was meant to be supported but just missing a piece, or not meant to be supported but very easy to add in. Finally, I was working from the 4.0 nightlies Debian repository, and the examples/scsf/kamailio.cfg file for 4.0.x seems to skip authentication. It looks like this section has been reworked heavily in 4.1.0, so this may be less important - but I put together a different kamailio.cfg based on a mailing list post (http://lists.sip-router.org/pipermail/sr-users/2013-March/077142.html), and it might make sense to use that as the default file for 4.0.x. The attached patches are against the 4.0 branch in Git (specifically, commit 1b98961522fd8a7eb73ecc7d1772541f8b81aabc). I'm happy to apply any feedback which more knowledgeable contributors have. Best regards, Rob

9 years, 2 months

Jump to page:

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

sr-dev December 2013