sr-dev October 2009

sr-dev@lists.kamailio.org

30 participants
468 discussions

by Henning Westerholt

Hi all, in the kamailio Makefile a "make distclean" actually cleans all modules, including the "not default" ones. In sr this not works, here make distclean only cleans the default modules. Was this done intentionally? If yes, how do i can clean this modules, "make distclean modules=modules/enum/" dont' work for example. Sorry if this was answered before, i did not found anything with a quick search - thanks, Henning

15 years, 1 month

CVS:commitlog: rtpproxy Makefile.in

by Maxim Sobolev

sobomax 2009/10/06 07:54:40 CEST SER CVS Repository Modified files: . Makefile.in Log: Regen. Revision Changes Path 1.15 +5 -3 rtpproxy/Makefile.in http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/Makefile.in.diff?r1=…

15 years, 1 month

CVS:commitlog: rtpproxy Makefile.am

by Maxim Sobolev

sobomax 2009/10/06 07:54:06 CEST SER CVS Repository Modified files: . Makefile.am Log: Connect rtpp_syslog_async.[ch] to the build. Revision Changes Path 1.16 +3 -2 rtpproxy/Makefile.am http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/Makefile.am.diff?r1=…

15 years, 1 month

CVS:commitlog: rtpproxy main.c rtpp_defines.h rtpp_log.c rtpp_syslog_async.c rtpp_syslog_async.h

by Maxim Sobolev

sobomax 2009/10/06 07:53:04 CEST SER CVS Repository Modified files: . main.c rtpp_defines.h rtpp_log.c Added files: . rtpp_syslog_async.c rtpp_syslog_async.h Log: Fix serious issue wich can happen with default loging method (syslog) at the default log level (debug). I'm unsure if it affects all Linux systems or not, but at least on one CentOS release 5.3 the vsyslog(3) call takes ridiculously large amount of time to complete. Few of such calls easily take 1-2 seconds. Since the RTPproxy is still single-threaded, this means that traffic halts for that duration, which is obviously bad. Run vsyslog(3) calls in the separate working thread instead. This ensures that logging won't interfere with normal activity even if the I/O system is loaded or when logging is done over network. While I am here, set default logging level when running in background to "WARN", since the current level ("DBUG") is too verbose to be a default. Revision Changes Path 1.93 +2 -2 rtpproxy/main.c http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/main.c.diff?r1=1.92&… 1.28 +1 -2 rtpproxy/rtpp_defines.h http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/rtpp_defines.h.diff?… 1.5 +26 -10 rtpproxy/rtpp_log.c http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/rtpp_log.c.diff?r1=1… 1.1 +241 -0 rtpproxy/rtpp_syslog_async.c (new) http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/rtpp_syslog_async.c?… 1.1 +38 -0 rtpproxy/rtpp_syslog_async.h (new) http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/rtpproxy/rtpp_syslog_async.h?…

15 years, 1 month

fr_timer

by Klaus Darilion

Hi! I just read the the documentation of the fr_timer nad I think it is wrong (as K's documentation is identical and wrong). http://sip-router.org/docbook/sip-router/branch/master/modules/tm/tm.html#f… AFAIK, fr_timer is triggered if there is no "provisional" response received, and not as the documentation said, a "final" response. Eg. I always used fr_timer: 2 seconds fr_inv_timer: 120 seconds Which means that the callee has to send a prov. response (e.g. 100 trying) within 2 seconds, to achieve fast failover. Once the prov. response is received, fr_timer is deactivated and fr_inv_timer is used. 1. Question: Is this the same in sr's tm module? If yes, then the docs should be fixed (I can do that). 2. If yes, what about renaming the parameters and introduce parameters which are clear, e.g.: timeout_prov_reply and timeout_final_reply regards klaus

15 years, 1 month

CVS:commitlog: sip_router sctp_server.c

by Andrei Pelinescu-Onciul

andrei 2009/10/05 17:28:06 CEST SER CVS Repository Modified files: . sctp_server.c Log: sctp: linux kernel obsolete api bug workaround set SPP_PMTUD_ENABLE when changing pathmaxrxt on linux. This was required on older versions of the api draft (<=14, most likely a draft bug) and is still required on linux. Reported-by: Libor Chocholaty libor at iptel org (cherry picked from commit 8459110abc3e0241e7b8a028e0252c97c8a5d9b5) (backported from sip-router) Revision Changes Path 1.33 +8 -1 sip_router/sctp_server.c http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/sip_router/sctp_server.c.diff…

15 years, 1 month

CVS:commitlog: sip_router tcp_main.c

by Andrei Pelinescu-Onciul

andrei 2009/10/05 17:23:08 CEST SER CVS Repository Modified files: . tcp_main.c Log: tcp: fix fd leak on async error or quick connect In some error cases (from which the most likely to appear in real-life would be maximum global send queue size exceeded), the fd for a new created connection (in tcp_send()) was not closed when the connection was destroyed. (backported from sip-router b8d528b6) Revision Changes Path 1.152 +4 -1 sip_router/tcp_main.c http://cvs.berlios.de/cgi-bin/viewcvs.cgi/ser/sip_router/tcp_main.c.diff?r1…

15 years, 1 month

git:master: sctp: linux kernel obsolete api bug workaround

by Andrei Pelinescu-Onciul

Module: sip-router Branch: master Commit: 8459110abc3e0241e7b8a028e0252c97c8a5d9b5 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=8459110… Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org> Date: Mon Oct 5 16:15:30 2009 +0200 sctp: linux kernel obsolete api bug workaround set SPP_PMTUD_ENABLE when changing pathmaxrxt on linux. This was required on older versions of the api draft (<=14, most likely a draft bug) and is still required on linux. Reported-by: Libor Chocholaty libor at iptel org --- sctp_server.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/sctp_server.c b/sctp_server.c index 909d735..9fa2b56 100644 --- a/sctp_server.c +++ b/sctp_server.c @@ -610,6 +610,13 @@ static int sctp_init_sock_opt_common(int s, int af) pp.spp_flags=SPP_HB_DISABLE; pp.spp_hbinterval=0; } +#ifdef __OS_linux + if (pp.spp_pathmaxrxt){ + /* hack to work on linux, pathmaxrxt is set only if + SPP_PMTUD_ENABLE */ + pp.spp_flags|=SPP_PMTUD_ENABLE; + } +#endif /*__OS_linux */ /* if at least one is non-null => we have to set it */ if (sctp_setsockopt(s, IPPROTO_SCTP, SCTP_PEER_ADDR_PARAMS, (void*)&pp, sizeof(pp), "setsockopt: SCTP_PEER_ADDR_PARAMS")!=0){

15 years, 1 month

make dbschema

by jh＠tutpro.com

henning, 'make dbschema' works, but takes way too much time. in k, the command completes in no time, but sr version takes forever. is sr version doing some extra stuff? -- juha

15 years, 1 month

SF.net SVN: openser:[5930] branches/1.5/modules/msilo/msilo.c

by Juha Heinanen

Revision: 5930 http://openser.svn.sourceforge.net/openser/?rev=5930&view=rev Author: juhe Date: 2009-10-05 12:20:50 +0000 (Mon, 05 Oct 2009) Log Message: ----------- * Send composed body only if content type is text/plain. Modified Paths: -------------- branches/1.5/modules/msilo/msilo.c This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

15 years, 1 month

← Newer
1
...
39
40
41
42
43
44
45
46
47
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

sr-dev October 2009