sr-dev June 2009

sr-dev@lists.kamailio.org

33 participants
284 discussions

SF.net SVN: openser:[5883] trunk/packaging/debian
by Inaki Baz Castillo 23 Jun '09

23 Jun '09

Revision: 5883 http://openser.svn.sourceforge.net/openser/?rev=5883&view=rev Author: ibc_sf Date: 2009-06-23 17:06:19 +0000 (Tue, 23 Jun 2009) Log Message: ----------- - Debian Control files: Replaced GnuTLS with OpenSSL since OpenSSL provides better performance: http://www.mail-archive.com/lftp-devel@uniyar.ac.ru/msg01487.html Modified Paths: -------------- trunk/packaging/debian/control trunk/packaging/debian-etch/control trunk/packaging/debian-lenny/control This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

1 0

Re: [sr-dev] Conflict between libcurl4-gnutls-dev and libcurl4-openssl-dev (Debian)
by Henning Westerholt 23 Jun '09

23 Jun '09

On Dienstag, 23. Juni 2009, you wrote: > > is it inherited in this case? the tls implementation in core of kamailio > > is binding directly to openssl library and therefore need for the gpl > > disclaimer, but in this case utils binds to curl library. > > > > In sip router this is fixed I guess and tls is shipped with debs -- > > anyhow, tls is a module in sr, the core is the same. Hi Jan, > If I understand it correctly, we are talking here about Debian control > files that are being kept in the git repository, right? But aren't those > files intended for people who want to build Debian packages themselves, > rather than for official Debian developers? As far as I know official > Debian developers maintain their control files elsewhere, and in fact it is > recommended that upstream projects do not provide their own control files. sure, our debian control files are different then the official ones. And they are maintained in a different repository, e.g. in "pkg-voip" for kamailio. But they are synchronised from time to time. > If this is the case, does it really matter if we keep libcurl4-openssl-dev > in our control files? We know that all the stuff should work (at least was > developed with) with openssl, so shouldn't this be the default for the > people who attempt to build the packages themselves? > > Also note that the -dev packages are only needed to actually build > packages, they are not needed at runtime. They only contain files that are > needed to compile applications that use libcurl. Corresponding runtime > libraries are in libcurl3 (openssl version) and libcurl3-gnutls (gnutls > version) and these can be installed at the same time. What this means is > that it is not possible to *compile* applications for use with > libcurl-openssl and libcurl-gnutls at the same time. I've nothing against it, just wanted to note that it makes packaging a bit harder from a debian POV. Henning

1 0

git:master: tmx: added K compatible tm statistics
by Daniel-Constantin Mierla 23 Jun '09

23 Jun '09

Module: sip-router Branch: master Commit: f01d1251f12325b777a7af980fa1d5c2be2caabf URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=f01d125… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: Tue Jun 23 12:46:21 2009 +0200 tmx: added K compatible tm statistics --- modules_k/tmx/tmx_mod.c | 125 ++++++++++++++++++++++++++++++++++++++++++---- 1 files changed, 114 insertions(+), 11 deletions(-) diff --git a/modules_k/tmx/tmx_mod.c b/modules_k/tmx/tmx_mod.c index 711e966..e06f505 100644 --- a/modules_k/tmx/tmx_mod.c +++ b/modules_k/tmx/tmx_mod.c @@ -61,18 +61,35 @@ stat_var *tm_trans_6xx; stat_var *tm_trans_inuse; #ifdef STATISTICS + +unsigned long tmx_stats_uas_trans(void); +unsigned long tmx_stats_uac_trans(void); +unsigned long tmx_stats_trans_2xx(void); +unsigned long tmx_stats_trans_3xx(void); +unsigned long tmx_stats_trans_4xx(void); +unsigned long tmx_stats_trans_5xx(void); +unsigned long tmx_stats_trans_6xx(void); +unsigned long tmx_stats_trans_inuse(void); +#if 0 +unsigned long tmx_stats_rcv_rpls(void); +unsigned long tmx_stats_rld_rpls(void); +#endif +unsigned long tmx_stats_loc_rpls(void); + static stat_export_t mod_stats[] = { - {"received_replies" , 0, &tm_rcv_rpls }, - {"relayed_replies" , 0, &tm_rld_rpls }, - {"local_replies" , 0, &tm_loc_rpls }, - {"UAS_transactions" , 0, &tm_uas_trans }, - {"UAC_transactions" , 0, &tm_uac_trans }, - {"2xx_transactions" , 0, &tm_trans_2xx }, - {"3xx_transactions" , 0, &tm_trans_3xx }, - {"4xx_transactions" , 0, &tm_trans_4xx }, - {"5xx_transactions" , 0, &tm_trans_5xx }, - {"6xx_transactions" , 0, &tm_trans_6xx }, - {"inuse_transactions" , STAT_NO_RESET, &tm_trans_inuse }, + {"UAS_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_uas_trans }, + {"UAC_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_uac_trans }, + {"2xx_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_2xx }, + {"3xx_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_3xx }, + {"4xx_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_4xx }, + {"5xx_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_5xx }, + {"6xx_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_6xx }, + {"inuse_transactions" , STAT_IS_FUNC, (stat_var**)tmx_stats_trans_inuse }, +#if 0 + {"received_replies" , STAT_IS_FUNC, (stat_var**)tmx_stats_rcv_rpls }, + {"relayed_replies" , STAT_IS_FUNC, (stat_var**)tmx_stats_rld_rpls }, +#endif + {"local_replies" , STAT_IS_FUNC, (stat_var**)tmx_stats_loc_rpls }, {0,0,0} }; #endif @@ -235,3 +252,89 @@ static int t_cancel_branches(struct sip_msg* msg, char *k, char *s2) return 1; } + +#ifdef STATISTICS + +/*** tm stats ***/ + +static struct t_proc_stats _tmx_stats_all; +static ticks_t _tmx_stats_tm = 0; +void tmx_stats_update(void) +{ + ticks_t t; + t = get_ticks(); + if(t!=_tmx_stats_tm) { + _tmx_tmb.get_stats(&_tmx_stats_all); + _tmx_stats_tm = t; + } +} + +unsigned long tmx_stats_uas_trans(void) +{ + tmx_stats_update(); + return _tmx_stats_all.transactions; +} + +unsigned long tmx_stats_uac_trans(void) +{ + tmx_stats_update(); + return _tmx_stats_all.client_transactions; +} + +unsigned long tmx_stats_trans_2xx(void) +{ + tmx_stats_update(); + return _tmx_stats_all.completed_2xx; +} + +unsigned long tmx_stats_trans_3xx(void) +{ + tmx_stats_update(); + return _tmx_stats_all.completed_3xx; +} + +unsigned long tmx_stats_trans_4xx(void) +{ + tmx_stats_update(); + return _tmx_stats_all.completed_4xx; +} + +unsigned long tmx_stats_trans_5xx(void) +{ + tmx_stats_update(); + return _tmx_stats_all.completed_5xx; +} + +unsigned long tmx_stats_trans_6xx(void) +{ + tmx_stats_update(); + return _tmx_stats_all.completed_6xx; +} + +unsigned long tmx_stats_trans_inuse(void) +{ + tmx_stats_update(); + return (_tmx_stats_all.transactions - _tmx_stats_all.deleted); +} + +#if 0 +unsigned long tmx_stats_rcv_rpls(void) +{ + tmx_stats_update(); + return 0; +} + +unsigned long tmx_stats_rld_rpls(void) +{ + tmx_stats_update(); + return 0; +} +#endif + +unsigned long tmx_stats_loc_rpls(void) +{ + tmx_stats_update(); + return _tmx_stats_all.replied_locally; +} + +#endif

1 0

git:master: tm: small wrapper to get TM stats via API
by Daniel-Constantin Mierla 23 Jun '09

23 Jun '09

Module: sip-router Branch: master Commit: a98476c55ffdce59bff96ac63d00fad0c337b70b URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=a98476c… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: Tue Jun 23 12:44:39 2009 +0200 tm: small wrapper to get TM stats via API - used for K style statistics --- modules/tm/t_stats.c | 13 +++++++++++++ modules/tm/t_stats.h | 3 +++ modules/tm/tm_load.c | 1 + modules/tm/tm_load.h | 2 ++ 4 files changed, 19 insertions(+), 0 deletions(-) diff --git a/modules/tm/t_stats.c b/modules/tm/t_stats.c index 9e2d192..627b379 100644 --- a/modules/tm/t_stats.c +++ b/modules/tm/t_stats.c @@ -175,6 +175,19 @@ void tm_rpc_stats(rpc_t* rpc, void* c) /* rpc->fault(c, 100, "Trying"); */ } +int tm_get_stats(struct t_proc_stats *all) +{ + int i, pno; + if(all==NULL) + return -1; + + pno = get_max_procs(); + memset(all, 0, sizeof(struct t_proc_stats)); + for(i = 0; i < pno; i++) { + tm_proc_stats_add(all, all, &tm_stats[i].s); + } + return 0; +} /* hash statistics */ diff --git a/modules/tm/t_stats.h b/modules/tm/t_stats.h index d38c751..734b4ad 100644 --- a/modules/tm/t_stats.h +++ b/modules/tm/t_stats.h @@ -147,4 +147,7 @@ void tm_rpc_stats(rpc_t* rpc, void* c); extern const char* tm_rpc_hash_stats_doc[2]; void tm_rpc_hash_stats(rpc_t* rpc, void* c); +typedef int (*tm_get_stats_f)(struct t_proc_stats *all); +int tm_get_stats(struct t_proc_stats *all); + #endif diff --git a/modules/tm/tm_load.c b/modules/tm/tm_load.c index 22e1782..5fa054a 100644 --- a/modules/tm/tm_load.c +++ b/modules/tm/tm_load.c @@ -235,6 +235,7 @@ int load_tm( struct tm_binds *tmb) tmb->t_check = t_check; tmb->unref_cell = unref_cell; tmb->which_cancel = which_cancel; + tmb->get_stats = tm_get_stats; #ifdef WITH_TM_CTX tmb->tm_ctx_get = tm_ctx_get; diff --git a/modules/tm/tm_load.h b/modules/tm/tm_load.h index 54c0b9a..2d5bf46 100644 --- a/modules/tm/tm_load.h +++ b/modules/tm/tm_load.h @@ -49,6 +49,7 @@ #include "callid.h" #include "t_cancel.h" #include "t_suspend.h" +#include "t_stats.h" /* export not usable from scripts */ #define NO_SCRIPT -1 @@ -142,6 +143,7 @@ struct tm_binds { tcheck_f t_check; unref_cell_f unref_cell; which_cancel_f which_cancel; + tm_get_stats_f get_stats; #ifdef WITH_TM_CTX tm_ctx_get_f tm_ctx_get; #else

1 0

git:master: nathelper(k): fixed content type check
by Daniel-Constantin Mierla 23 Jun '09

23 Jun '09

Module: sip-router Branch: master Commit: a0993fbf1d35b27644c6837eff3832c1e12d2495 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=a0993fb… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: Tue Jun 23 11:58:26 2009 +0200 nathelper(k): fixed content type check - mistakenly introduced with multipart body support - reported by Pascal Maugeri --- modules_k/nathelper/nhelpr_funcs.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/modules_k/nathelper/nhelpr_funcs.c b/modules_k/nathelper/nhelpr_funcs.c index 95f2133..50ec7d5 100644 --- a/modules_k/nathelper/nhelpr_funcs.c +++ b/modules_k/nathelper/nhelpr_funcs.c @@ -200,7 +200,7 @@ int extract_body(struct sip_msg *msg, str *body ) /* no need for parse_headers(msg, EOH), get_body will * parse everything */ /*is the content type correct?*/ - if((ret = check_content_type(msg))==1) + if((ret = check_content_type(msg))==-1) { LM_ERR("content type mismatching\n"); goto error;

1 0

tm error: status rewrite by UAS
by jh＠tutpro.com 22 Jun '09

22 Jun '09

while testing new version of lcr module, i got this kind of tm errors: Jun 22 11:27:36 localhost /usr/sbin/sip-router[16330]: ERROR: tm [t_reply.c:1066]: ERROR: t_should_relay_response: status rewrite by UAS: stored: 408, received: 403 Jun 22 11:27:36 localhost /usr/sbin/sip-router[16329]: ERROR: tm [t_reply.c:1066]: ERROR: t_should_relay_response: status rewrite by UAS: stored: 408, received: 403 Jun 22 11:27:37 localhost /usr/sbin/sip-router[16331]: ERROR: tm [t_reply.c:1066]: ERROR: t_should_relay_response: status rewrite by UAS: stored: 408, received: 403 Jun 22 11:27:39 localhost /usr/sbin/sip-router[16331]: ERROR: tm [t_reply.c:1066]: ERROR: t_should_relay_response: status rewrite by UAS: stored: 408, received: 403 they came when i first called t_relay 4 times. first call was in route block and the rest in failure route block. there was no reply to any of these four calls and each time fr_timer hit. on fifth call, gw responded with 403. i don't get these kind of errors in k. what do they mean and how to get rid of them? -- juha

3 9

git:master: tm doc: t_check_trans and note about fr_timer_avp
by Andrei Pelinescu-Onciul 22 Jun '09

22 Jun '09

Module: sip-router Branch: master Commit: c263b1a7a83f7143093e699754b377647d54496d URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=c263b1a… Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Date: Mon Jun 22 17:26:47 2009 +0200 tm doc: t_check_trans and note about fr_timer_avp - added missing t_check_trans docs - added fr_timer_avp and fr_inv_timer_avp notes about new empty/zero avp value handling (use default value) and about using seconds instead of milliseconds. --- modules/tm/doc/functions.xml | 75 ++++++++++++++++++++++++++++++++++++++++++ modules/tm/doc/params.xml | 30 ++++++++++++++-- 2 files changed, 101 insertions(+), 4 deletions(-) diff --git a/modules/tm/doc/functions.xml b/modules/tm/doc/functions.xml index 70309a5..d3b21d0 100644 --- a/modules/tm/doc/functions.xml +++ b/modules/tm/doc/functions.xml @@ -1104,4 +1104,79 @@ if (!t_next_contacts()) { </example> </section> + <section id="t_check_trans"> + <title> + <function>t_check_trans()</function> + </title> + <para> + <function>t_check_trans()</function> can be used to quickly check if + a message belongs or is related to a transaction. It behaves + differently for different types of messages: + <itemizedlist> + <listitem> + <para>For a SIP Reply it returns true if the reply belongs to + an existing transaction and false otherwise.</para> + </listitem> + <listitem> + <para>For a CANCEL it behaves exactly as + <function>t_lookup_cancel()</function>: returns true if a + corresponding INVITE transaction exists for the CANCEL and + false otherwise.</para> + </listitem> + <listitem> + <para>For ACKs to negative replies or for ACKs to local + transactions it will terminate the script if the ACK belongs + to a transaction (it would make very little sense to process + an ACK to a negative reply for an existing transaction in + some other way then to simply pass it to tm) or return false + if not.</para> + </listitem> + <listitem> + <para>For end-to-end ACKs (ACKs to 2xx responses for forwarded + INVITE transactions) it will return true if the corresponding + INVITE transaction is found and still active and false if not. + </para> + <note>Note that the e2e ACK matching is more of a hint + then a certainty. A delayed e2e ACK might arrive after the + transaction wait time elapses, when the INVITE transaction no + longer exists and thus would not match anything. There are + also cases when tm would not keep all the information needed + for e2e ACK matching (since this is not needed for a statefull + proxy and it requires additional memory, tm will not keep this + information unless needed by some other module or callbacks). + </note> + </listitem> + <listitem> + <para>For other requests (non ACKs and non CANCELs), it will + terminate the script for retransmissions and return false for + new requests (for which no transaction exists yet).</para> + </listitem> + </itemizedlist> + </para> + <note><para> + An important difference from kamailio version is that for an ACK to + negative reply or for a local transaction, the script execution will be + immediately stopped and the message handled by tm, instead of returning + true. + </para></note> + <para><function>t_check_trans()</function> functionality for requests, + except for the e2e ACK matching, can be replicated in the script + using <function>t_lookup_cancel()</function> and + <function>t_lookup_request()</function>. + </para> + <para>See also: <function>t_lookup_request()</function>, + <function>t_lookup_cancel()</function>. + </para> + <example> + <title><function>t_check_trans</function> usage</title> + <programlisting> + +if ( method == "CANCEL" && !t_check_trans()) + sl_reply("403", "cancel out of the blue forbidden"); +# note: in this example t_check_trans() can be replaced by t_lookup_cancel() + + </programlisting> + </example> + </section> + </section> diff --git a/modules/tm/doc/params.xml b/modules/tm/doc/params.xml index ddfcb29..7df9f13 100644 --- a/modules/tm/doc/params.xml +++ b/modules/tm/doc/params.xml @@ -809,6 +809,17 @@ modparam("tm", "contacts_avp", "$avp(i:25)") The value of this parameter is the the name of the AVP to be checked, without the $ character or "$avp" prefix. </para> + <note><para> + The value of the AVP is expected to be expressed in + <emphasis>seconds</emphasis> and not milliseconds (unlike the rest + of the timers). + </para></note> + <para> + This parameter is kept for backwards compatibility (hence its + value expressed in seconds instead of milliseconds and its arcane + way of specifying the avps). The recommended replacement is using + <function>t_set_fr()</function> on a per transaction basis. + </para> <para> See also: <function>t_set_fr()</function>, <varname>fr_timer</varname>. @@ -830,15 +841,26 @@ modparam("tm", "fr_timer_avp", "i:708") per-transaction basis. The administrator can provide a value to be used for a particular transaction in an AVP. This parameter contains the name of the AVP that will be checked. If the AVP - exists then its value will be used for the fr_inv_timer timer, - effectively overriding the value configured in - <varname>fr_inv_timer</varname> parameter for the current - transaction. + exists, is non-empty and non-zero then its value will be used + for the fr_inv_timer timer, effectively overriding the value + configured in <varname>fr_inv_timer</varname> parameter for the + current transaction. </para> <para> The value of this parameter is the the name of the AVP to be checked, without the $ character or "$avp" prefix. </para> + <note><para> + The value of the AVP is expected to be expressed in + <emphasis>seconds</emphasis> and not milliseconds (unlike the rest + of the timers). + </para></note> + <para> + This parameter is kept for backwards compatibility (hence its + value expressed in seconds instead of milliseconds and its arcane + way of specifying the avps). The recommended replacement is using + <function>t_set_fr()</function> on a per transaction basis. + </para> <para> See also: <function>t_set_fr()</function>, <varname>fr_inv_timer</varname>.

1 0

git:master: tm doc: README re-generated
by Andrei Pelinescu-Onciul 22 Jun '09

22 Jun '09

Module: sip-router Branch: master Commit: e5cf7a1831e380cd6bd43e5a36dfd6d04625bc41 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=e5cf7a1… Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Date: Mon Jun 22 17:32:44 2009 +0200 tm doc: README re-generated --- modules/tm/README | 168 ++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 files changed, 165 insertions(+), 3 deletions(-) diff --git a/modules/tm/README b/modules/tm/README index db20c8d..39e3261 100644 --- a/modules/tm/README +++ b/modules/tm/README @@ -15,6 +15,97 @@ Juha Heinanen Revision $Revision$ $Date$ __________________________________________________________________ + 1.1. Overview + 1.2. Known Issues + 1.3. Parameters + + 1.3.1. fr_timer (integer) + 1.3.2. fr_inv_timer (integer) + 1.3.3. max_inv_lifetime (integer) + 1.3.4. max_noninv_lifetime (integer) + 1.3.5. wt_timer (integer) + 1.3.6. delete_timer (integer) + 1.3.7. retr_timer1 (integer) + 1.3.8. retr_timer2 (integer) + 1.3.9. noisy_ctimer (integer) + 1.3.10. restart_fr_on_each_reply (integer) + 1.3.11. auto_inv_100 (integer) + 1.3.12. auto_inv_100_reason (string) + 1.3.13. unix_tx_timeout (integer) + 1.3.14. aggregate_challenges (integer) + 1.3.15. reparse_invite (integer) + 1.3.16. ac_extra_hdrs (string) + 1.3.17. blst_503 (integer) + 1.3.18. blst_503_def_timeout (integer) + 1.3.19. blst_503_min_timeout (integer) + 1.3.20. blst_503_max_timeout (integer) + 1.3.21. blst_methods_add (unsigned integer) + 1.3.22. blst_methods_lookup (unsigned integer) + 1.3.23. cancel_b_method (integer) + 1.3.24. reparse_on_dns_failover (integer) + 1.3.25. on_sl_reply (string) + 1.3.26. fr_inv_timer_next (integer) + 1.3.27. contacts_avp (string) + 1.3.28. fr_timer_avp (string) + 1.3.29. fr_inv_timer_avp (string) + + 1.4. Functions + + 1.4.1. t_relay_to_udp(ip, port), t_relay_to_udp(), + t_relay_to_tcp(ip, port) t_relay_to_tcp() + t_relay_to_tls(ip, port) t_relay_to_tls() + t_relay_to_sctp(ip, port) t_relay_to_sctp() + + 1.4.2. t_relay() t_relay(host, port) + 1.4.3. t_on_failure(failure_route) + 1.4.4. t_on_reply(onreply_route) + 1.4.5. t_on_branch(branch_route) + 1.4.6. append_branch() + 1.4.7. t_newtran() + 1.4.8. t_reply(code, reason_phrase) + 1.4.9. t_lookup_request() + 1.4.10. t_retransmit_reply() + 1.4.11. t_release() + 1.4.12. t_forward_nonack() t_forward_nonack(ip, port) + t_forward_nonack_udp(ip, port) t_forward_nonack_tcp(ip, + port) t_forward_nonack_tls(ip, port) + t_forward_nonack_sctp(ip, port) + + 1.4.13. t_set_fr(fr_inv_timeout [, fr_timeout]) + 1.4.14. t_reset_fr() + 1.4.15. t_set_max_lifetime(inv_lifetime, noninv_lifetime) + 1.4.16. t_reset_max_lifetime() + 1.4.17. t_set_retr(retr_t1_interval, retr_t2_interval) + 1.4.18. t_reset_retr() + 1.4.19. t_set_auto_inv_100(0|1) + 1.4.20. t_branch_timeout() + 1.4.21. t_branch_replied() + 1.4.22. t_any_timeout() + 1.4.23. t_any_replied() + 1.4.24. t_grep_status("code") + 1.4.25. t_is_canceled() + 1.4.26. t_is_expired() + 1.4.27. t_relay_cancel() + 1.4.28. t_lookup_cancel(), t_lookup_cancel(1) + 1.4.29. t_drop_replies() + 1.4.30. t_save_lumps() + 1.4.31. t_load_contacts() + 1.4.32. t_next_contacts() + 1.4.33. t_check_trans() + + 1.5. TM Module API + + 1.5.1. Defines + 1.5.2. Functions + + 1.5.2.1. register_tmcb(cb_type, cb_func) + 1.5.2.2. load_tm(*import_structure) + 1.5.2.3. int t_suspend(struct sip_msg *msg, unsigned int + *hash_index, unsigned int *label) + + 1.5.2.4. int t_continue(unsigned int hash_index, unsigned + int label, struct action *route) + 1.1. Overview TM module enables stateful processing of SIP transactions. The main use @@ -617,6 +708,16 @@ modparam("tm", "contacts_avp", "$avp(i:25)") The value of this parameter is the the name of the AVP to be checked, without the $ character or "$avp" prefix. +Note + + The value of the AVP is expected to be expressed in seconds and not + milliseconds (unlike the rest of the timers). + + This parameter is kept for backwards compatibility (hence its value + expressed in seconds instead of milliseconds and its arcane way of + specifying the avps). The recommended replacement is using t_set_fr() + on a per transaction basis. + See also: t_set_fr(), fr_timer. Example 28. Set fr_timer_avp parameter @@ -629,13 +730,24 @@ modparam("tm", "fr_timer_avp", "i:708") The value of fr_inv_timer timer can be overriden on per-transaction basis. The administrator can provide a value to be used for a particular transaction in an AVP. This parameter contains the name of - the AVP that will be checked. If the AVP exists then its value will be - used for the fr_inv_timer timer, effectively overriding the value - configured in fr_inv_timer parameter for the current transaction. + the AVP that will be checked. If the AVP exists, is non-empty and + non-zero then its value will be used for the fr_inv_timer timer, + effectively overriding the value configured in fr_inv_timer parameter + for the current transaction. The value of this parameter is the the name of the AVP to be checked, without the $ character or "$avp" prefix. +Note + + The value of the AVP is expected to be expressed in seconds and not + milliseconds (unlike the rest of the timers). + + This parameter is kept for backwards compatibility (hence its value + expressed in seconds instead of milliseconds and its arcane way of + specifying the avps). The recommended replacement is using t_set_fr() + on a per transaction basis. + See also: t_set_fr(), fr_inv_timer. Example 29. Set fr_inv_timer_avp parameter @@ -1350,6 +1462,56 @@ if (!t_next_contacts()) { }; ... +1.4.33. t_check_trans() + + t_check_trans() can be used to quickly check if a message belongs or is + related to a transaction. It behaves differently for different types of + messages: + * For a SIP Reply it returns true if the reply belongs to an existing + transaction and false otherwise. + * For a CANCEL it behaves exactly as t_lookup_cancel(): returns true + if a corresponding INVITE transaction exists for the CANCEL and + false otherwise. + * For ACKs to negative replies or for ACKs to local transactions it + will terminate the script if the ACK belongs to a transaction (it + would make very little sense to process an ACK to a negative reply + for an existing transaction in some other way then to simply pass + it to tm) or return false if not. + * For end-to-end ACKs (ACKs to 2xx responses for forwarded INVITE + transactions) it will return true if the corresponding INVITE + transaction is found and still active and false if not. + +Note + Note that the e2e ACK matching is more of a hint then a certainty. + A delayed e2e ACK might arrive after the transaction wait time + elapses, when the INVITE transaction no longer exists and thus + would not match anything. There are also cases when tm would not + keep all the information needed for e2e ACK matching (since this is + not needed for a statefull proxy and it requires additional memory, + tm will not keep this information unless needed by some other + module or callbacks). + * For other requests (non ACKs and non CANCELs), it will terminate + the script for retransmissions and return false for new requests + (for which no transaction exists yet). + +Note + + An important difference from kamailio version is that for an ACK to + negative reply or for a local transaction, the script execution will be + immediately stopped and the message handled by tm, instead of returning + true. + + t_check_trans() functionality for requests, except for the e2e ACK + matching, can be replicated in the script using t_lookup_cancel() and + t_lookup_request(). + + See also: t_lookup_request(), t_lookup_cancel(). + + Example 62. t_check_trans usage +if ( method == "CANCEL" && !t_check_trans()) + sl_reply("403", "cancel out of the blue forbidden"); +# note: in this example t_check_trans() can be replaced by t_lookup_cancel() + 1.5. TM Module API Revision History

1 0

git:master: tm: ignore avp timers with 0 or empty value
by Andrei Pelinescu-Onciul 22 Jun '09

22 Jun '09

Module: sip-router Branch: master Commit: bcc993680ca2722aec2e4f240569cb7979808e18 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=bcc9936… Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Date: Mon Jun 22 16:00:10 2009 +0200 tm: ignore avp timers with 0 or empty value Kamailio compatibility: If an avp timer (fr_timer_avp or fr_inv_timer_avp) is empty ("") or 0, use the default timer value (fr_timer or fr_inv_timer). Reported-by: Juha Heinanen jh at tutpro com. --- modules/tm/t_funcs.c | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/modules/tm/t_funcs.c b/modules/tm/t_funcs.c index c228b60..d60c99e 100644 --- a/modules/tm/t_funcs.c +++ b/modules/tm/t_funcs.c @@ -416,8 +416,9 @@ int init_avp_params(char *fr_timer_param, char *fr_inv_timer_param, } -/* - * Get the FR_{INV}_TIMER from corresponding AVP +/** Get the FR_{INV}_TIMER from corresponding AVP. + * @return 0 on success (use *timer) or 1 on failure (avp non-existent, + * avp present but empty/0, avp value not numeric). */ static inline int avp2timer(unsigned int* timer, int type, int_str name) { @@ -443,7 +444,7 @@ static inline int avp2timer(unsigned int* timer, int type, int_str name) *timer = val_istr.n; } - return 0; + return *timer==0; /* 1 if 0 (use default), 0 if !=0 (use *timer) */ }

1 0

git:master: nathelper(k): skip trailing whitespaces and crlf
by Daniel-Constantin Mierla 22 Jun '09

22 Jun '09

Module: sip-router Branch: master Commit: caef5ace2026fcb2a9dffb471e2cec6d843082f8 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=caef5ac… Author: Daniel-Constantin Mierla <miconda(a)gmail.com> Committer: Daniel-Constantin Mierla <miconda(a)gmail.com> Date: Mon Jun 22 17:25:18 2009 +0200 nathelper(k): skip trailing whitespaces and crlf - was affecting multipart bodies, causing nortpproxy string to be placed in the wrong position - reported by Pascal Maugeri --- modules_k/nathelper/nhelpr_funcs.c | 33 ++++++++++++++++++--------------- 1 files changed, 18 insertions(+), 15 deletions(-) diff --git a/modules_k/nathelper/nhelpr_funcs.c b/modules_k/nathelper/nhelpr_funcs.c index 4d5bd99..95f2133 100644 --- a/modules_k/nathelper/nhelpr_funcs.c +++ b/modules_k/nathelper/nhelpr_funcs.c @@ -206,21 +206,9 @@ int extract_body(struct sip_msg *msg, str *body ) goto error; } - for (skip = 0; skip < body->len; skip++) { - c = body->s[body->len - skip - 1]; - if (c != '\r' && c != '\n') - break; - } - if (skip == body->len) { - LM_ERR("empty body"); - goto error; - } - body->len -= skip; - - /*LM_DBG("DEBUG:extract_body:=|%.*s|\n",body->len,body->s);*/ - if(ret!=2) - return 1; + goto done; + /* multipart body */ if(get_mixed_part_delimiter(&msg->content_type->body,&mpdel) < 0) { goto error; @@ -266,12 +254,27 @@ int extract_body(struct sip_msg *msg, str *body ) { body->s = rest; body->len = p2-rest; - return 1; + goto done; } } error: return -1; + +done: + for (skip = 0; skip < body->len; skip++) { + c = body->s[body->len - skip - 1]; + if (c != '\r' && c != '\n') + break; + } + if (skip == body->len) { + LM_ERR("empty body"); + goto error; + } + body->len -= skip; + + /*LM_DBG("DEBUG:extract_body:=|%.*s|\n",body->len,body->s);*/ + return 1; } /*

1 0

← Newer
1
...
10
11
12
13
14
15
16
...
29
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

sr-dev June 2009