sr-dev April 2014

sr-dev@lists.kamailio.org

26 participants
304 discussions

Outbound flow failure handling merged to master
by Hugh Waite 23 Nov '16

23 Nov '16

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.

4 11

TM possible deadlock
by Jason Penton 09 Apr '15

09 Apr '15

Hi All, I have been experiencing a deadlock when a timeout occurs on a t_relayed() INVITE. Going through the code I have noticed a possible chance of deadlock (without re-entrant enabled). Here is my thinking: t_should_relay_response() is called with REPLY_LOCK when the timer process fires on the fr_inv_timer (no response from the INVITE that was relayed, other than 100 provisional) and a 408 is generated. However, from within that function there are calls to run_failure_handlers() which in turn *could* try and lock the reply (viz. somebody having a t_reply() call in the cfg file - in failure route block). This would result in another lock on the same transaction's REPLY_LOCK.... Has anybody else experienced something like this? this is on master btw. Cheers Jason

2 16

Fw: query related to NOTIFY message size
by Piyush.Bansal＠relianceada.com 17 Oct '14

17 Oct '14

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.

4 10

RFC: updating default values
by Daniel-Constantin Mierla 02 Oct '14

02 Oct '14

Hello, I am opening this discussion to decide if there is need to adjust some of the default values we have in Kamailio. Many of them were set even like 10 years ago, so they might not be very actual anymore. The main goal is to get the best possible settings for common usage. To start with, here are some values that should be reviewed: - default private memory is 4MB - if the config is not that small, it might not be sufficient free pkg to play with (e.g., for sql_query() result, storage of $var(...) values). Should it be increased to 8MB or other value? Debian/Centos have startup script that sets its value to 8 via command line - default shared memory is 32MB - for a decent deployment with tm, location, lcr/dispatcher, permissions, and anti-flood, it might leave not much free space. Should we double it or set to a different value - usrloc - db_ops_ruid should enabled (1) - seems stable, without it there are problems updating/deleting location records when UA changes the call-id for same contact address. - usrloc - hash_size - now is 9, which results in 512 slots, meaning is ok for few thousands of registered users, for more, performance will decrease when doing save/lookup location -- should it be made 10 (1024 slots for internal hash table)? - auth_db - load_credentials defaults to 'rpid', meaning that the query to get the password will retrieve also the rpid column. I haven't see rpid being used that much lately (replaced by PAI/PPI). I would make this parameter empty by default to avoid querying for an extra column that is likely to be empty. Perhaps there are more, I just wanted to get started. Reply with your comments to above list as well as add new items you thinks their default values should be adjusted. Cheers, Daniel -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda

4 7

Kamailio on OS/X Mavericks
by Olle E. Johansson 04 Aug '14

04 Aug '14

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

3 6

[PATCH 1/1] websocket: remove libunistring dependency
by Timo Teräs 17 Jul '14

17 Jul '14

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

3 10

Using t_suspend()/t_continue() multiple times on the same transaction
by Peter Dunkley 02 Jul '14

02 Jul '14

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

5 12

git:master: kex: new stats RPC commands
by Ovidiu Sas 19 May '14

19 May '14

Module: sip-router Branch: master Commit: 6f685460f1d7fd6ff07e55f2a807322168638988 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=6f68546… Author: Ovidiu Sas <osas(a)voipembedded.com> Committer: Ovidiu Sas <osas(a)voipembedded.com> Date: Thu Jan 16 14:46:32 2014 -0500 kex: new stats RPC commands --- modules/kex/core_stats.c | 285 +++++++++++++++++++++++++++++++++++++++++ modules/kex/doc/kex.xml | 14 ++ modules/kex/doc/kex_admin.xml | 50 +++++++ 3 files changed, 349 insertions(+), 0 deletions(-) Diff: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commitdiff;h=6f6…

3 5

patch: utils - http_query() - more logging and return body also on 4xx
by Mikko Lehto 08 May '14

08 May '14

Hi Please find attached silly little patch to utils module. It modifies http_query() to log little bit more when curling fails and returns response body also for 4xx respose. -- Mikko Lehto

2 2

module writing & per message variable within my module
by jay binks 08 May '14

08 May '14

So Im writing a custom module to implement some of our business logic, there are a few steps that im building into the module, for example conform_callerid conform_destination what I want to achive should be fairly simple, but let me explain and im sure someone will tell me where to look. when I call conform_callerid, I want to store a few bits of info so Initially I store them in a global variable ( yup lazy & dangerous I guess ) then I write the pv "get" function, to return the value from the global. this was only done as a quick proof of concept. but being global, im sure the variable will be overwritten by other message etc etc.. so I effectively want to store some custom fields against the message. I didnt see this on the msg structure, but maybe I missed something or is there a better way to achieve this ?? basically I want to be able to call conform_callerid then when I call conform_destination I want to look at some fields from the conform_callerid process ( within the context of the message we are processing ) any pointers would be appreciated. -- Sincerely Jay

2 6

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

sr-dev April 2014