sr-users

sr-users@lists.kamailio.org

42 participants
37519 discussions

async_workers vs. async workers modparam in case of async_route()
by Alex Balashov 22 Nov '16

22 Nov '16

Hi, If I am using async_route() with delay, what is the relationship between the core async_workers parameter value and the async module 'workers' modparam? This doc change was intended to clarify it, I think: https://www.mail-archive.com/sr-dev@lists.sip-router.org/msg25002.html But it does not for me: > Number of worker processes to be started to handle the > asynchronous tasks for async_route() and async_sleep(). So, does that mean the module starts its own processes for these cases? If so, does that mean I don't need core async_workers at all? Or do I? If I do, what happens if I have fewer core async_workers than modparam workers? What if I have more? Should I aim to keep them in sync? Thanks, -- Alex -- Alex Balashov | Principal | Evariste Systems LLC Tel: +1-706-510-6800 (direct) / +1-800-250-5920 (toll-free) Web: http://www.evaristesys.com/, http://www.csrpswitch.com/

2 3

log_prefix ignored in resumed routes
by Alex Balashov 22 Nov '16

22 Nov '16

Hi, I have this log_prefix: log_prefix="[$ci $mt] " Which means xlog("L_INFO", "xyz") normally results in log entries like this: Nov 22 00:29:50 allegro-4.evaristesys.com /usr/local/sbin/kamailio[27625]: [1-27702(a)10.150.20.6 1] INFO: xyz But if xlog("L_INFO", "xyz") happens to occur in a route[] block that is resumed after async_route()—I have not tested other async functions—there is no prefix: Nov 22 00:29:52 allegro-4.evaristesys.com /usr/local/sbin/kamailio[27678]: INFO: xyz -- Alex -- Alex Balashov | Principal | Evariste Systems LLC Tel: +1-706-510-6800 (direct) / +1-800-250-5920 (toll-free) Web: http://www.evaristesys.com/, http://www.csrpswitch.com/

2 1

flags documentation
by Alex Balashov 22 Nov '16

22 Nov '16

Hi, I have this set up in my configuration boilerplate: -- flags DUMMY : 1, PDD_DAMPEN : 2, PROXY_MEDIA_SET : 3, #!ifdef WITH_NAT_TRAVERSAL SIPPING : 4, #!endif DUMMY2 : 12 -- I then use these aliases in transaction flags, e.g. setflag(PDD_DAMPEN). But I can honestly say I don't remember how I arrived at this approach. I wrote it somewhere in early 3.x days. However, when I look at the core documentation for the 'flags' declaration, all I find is: http://www.kamailio.org/wiki/cookbooks/4.4.x/core#flags Or namely: -- flags Alias name: bool -- And there is no entry for "bool". So, I couldn't have possibly arrived at this usage from the documentation. Can anyone help me reverse-engineer this historical phenomenon? And, given that the documentation doesn't support it particularly, is it still an encouraged practice? Or should one use #!defined constants instead, as in the stock config? Thanks, -- Alex -- Alex Balashov | Principal | Evariste Systems LLC Tel: +1-706-510-6800 (direct) / +1-800-250-5920 (toll-free) Web: http://www.evaristesys.com/, http://www.csrpswitch.com/

2 3

Message flags preservation for replies in onsend_route
by Alex Balashov 22 Nov '16

22 Nov '16

Hi, I have -- flags BLAH : 1, DUMMY : 12 onsend_route_reply=yes route { ... setflag(BLAH); if(!t_relay()) sl_reply_error(); } onsend_route { if(is_reply() && isflagset(BLAH)) do_things(); } -- My discovery has been that isflagset(BLAH) always == FALSE _unless_ an onreply_route has been armed for the request, regardless of whether it is necessary. In other words, the following causes isflagset(BLAH) to == TRUE (if set, when appropriate) in the onsend_route for the reply: -- flags BLAH : 1, DUMMY : 12 onsend_route_reply=yes onreply_route[BOGUS] { exit; } route { ... setflag(BLAH); t_on_reply("BOGUS"); if(!t_relay()) sl_reply_error(); } onsend_route { if(is_reply() && isflagset(BLAH)) do_things(); } -- But without the onreply_route, it doesn't work. -- Alex Balashov | Principal | Evariste Systems LLC Tel: +1-706-510-6800 (direct) / +1-800-250-5920 (toll-free) Web: http://www.evaristesys.com/, http://www.csrpswitch.com/

1 0

Registering Trunk help please
by Andrei Mahalean 22 Nov '16

22 Nov '16

Hello, I am having a bit of trouble finding information on how to setup a registering trunk with Kamailio. By registering trunk, I mean a trunk between a PBX and Kamailio, of say 15 DID's, where the main DID (Pilot) registers with Kamailio and all the other 14 DID are able to make/receive calls without having to register all DID's individually. If the pilot DID registration goes away, the trunk is then considered to be unreachable/down. I have seen this in the past with a Broadsoft setup, but I can't figure out a way to do it with Kamailio. I have my pilot DID setup and registered with Kamailio ok and I can make calls in and out through my PSTN gateway, but I'm unsure how to group all the other DID's so they are matched against the pilot DID. Is this possible? Any help is appreciated. Regards, Andrei. -- Regards, Andrei.

2 1

Roadmap to Kamailio v5.0
by Daniel-Constantin Mierla 21 Nov '16

21 Nov '16

Hello, following the last IRC devel meeting, I am writing the summary of the main points being discussed there and what's planned to get to Kamailio v5.0. 1) Freezing should be done either before Chrismas holidays or at beginning of January 2017 to aim meeting the deadline for next Debian stable distro 2) Restructuring the source tree To be done in the weeks before freezing, so eventual fixes won't have hard time to be backported to 4.4. This means it will be done in December. Open issues: a) anyone aware of a tool that can update include directives upon relocation of the file? b) who is interested to join an eventual online session to perform the restructuring? C coding is not a must here, but more knowledge of sed/awk/perl/python/... and Makefiles ... 3) Defining a template for generating main file of a module The scope is to have a descriptive file for the module where one lists exported symbols (config params, functions, kemi functions, rpc/mi commands, etc.) which along with a template will generate the c code. The effective implementation of the functions/commands will be done separately, in a different file (e.g, the main file that is generated can have references to parameter variables with extern ... and will also generate the .h which will have only the prototypes for the functions). The main goal here is to get a way of easily generated an index of exported symbols that can be used to discover missing documentation. Right now we sort of do 'gentlemen agreement' on documenting functions and parameters, but many rpc commands are not documented (they are extracted with some tool scanning compiled objects). Also, same concept as with c code can be used to generate some stubs for documentation that can be included in docbook xml files. Few proposals of templating engines were mentioned, coming in my mind now being jinja2 and mustache. Anyone else with more suggestions? Or even propose alternatives of what can be used to solve this relation between code and docs? Cheers, Daniel -- Daniel-Constantin Mierla http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio Advanced Training, Berlin, Nov 28-30, 2016 - http://www.asipto.com

1 0

usage of app_mono : how to communicate results back to config file
by Pranathi Venkatayogi 21 Nov '16

21 Nov '16

Hi, I am using app_mono module and building my logic in c# app. I am unable to understand how to send the decision back to kamalio.cfg. More precisely: I would like my routing logic to allow specific user names though they are not registered as subscribers. Like: translate(a)myip.cyracom.com<mailto:translate@myip.cyracom.com> I have the logic in c# app which takes $ru and validates such that: Host is myip.cyracom.com User is translate How do I communicate the decision if it is valid or not back to my config script. Do I need to introduce new pseudo variable for this purpose? Thanks, Pranathi

2 1

Is there a way to detect inactive MSRP chat sessions without using SIP session timers
by Pranathi Venkatayogi 18 Nov '16

18 Nov '16

Hi, I am using MSRP feature of Blink client. I want to track when the chat starts and ends so I can free up server resources. Apparently Blink client does not send BYE when MSRP chat window closes. It does not support SIP session timers even. So potential use of SIP session timers (sst and dialog modules) is ruled out. Only possible way is to track every message that goes through the server and keep the session alive. If no message comes within “expiry time period”, server can send “bye” to all parties and tear up the session. Can you please comment if there are other approaches to solve this problem? If not, are they any modules that will help me implement the above approach easily? Thanks Pranathi

2 2

tcp_set_connection_lifetime - does not close the connection immediately
by Andy 18 Nov '16

18 Nov '16

I am calling tcp_set_connection_lifetime(conid, 1), which I understand is one sec after now, but I see it taking upto 60 secs to kill the connection, is there any other timer which has a larger interval which might affect this. -- View this message in context: http://sip-router.1086192.n5.nabble.com/tcp-set-connection-lifetime-does-no… Sent from the Users mailing list archive at Nabble.com.

2 2

Htable sht_rm_name_re usage
by Grant Bagdasarian 18 Nov '16

18 Nov '16

Hello Community, We were facing a weird issue in one of our Kamailio servers where custom htable entries were removed which did not belong to a particular call. Background Information: We use htable to store the call-id and a uuid when an INVITE message is received and remove these entries when the call is completed/failed (so on CANCEL, BYE, 4XX, 5XX, 6XX, etc). In this case the function sht_rm_name_re was causing incremented call-id's belonging to different calls to be removed from the htable, since the removal is based on regex and not the exact key. On INVITE we do this: $var(reference) = $uuid(g); $sht(calls=>$ci) = $ci; $sht(calls=>$ci::reference) = $var(reference); On BYE/4XX,5XX,6XX, etc we did this: sht_rm_name_re("calls=>$ci"); sht_rm_name_re("calls=>$ci::reference"); SIPP generates incremented call-id's, like this: 1-14875(a)1.0.0.1<mailto:1-14875@1.0.0.1>. When you generate enough calls, for instance 30 calls, the 21st call will have the call-id 21-14875(a)1.0.0.1<mailto:21-14875@1.0.0.1>. If for instance the BYE for the call-id 1-14875(a)1.0.0.1<mailto:1-14875@1.0.0.1> arrives before 21-14875(a)1.0.0.1<mailto:21-14875@1.0.0.1> , both htable entries will be removed by the sht_rm_name_re function, since it matches based on regex. To fix this we made sure the function matches on the exact keys by doing this: $var(callid) = '^' + $ci + '$'; $var(callid_reference) = '^' + $ci + '::reference' + '$'; sht_rm_name_re("calls=>$var(callid)"); sht_rm_name_re("calls=>$var(callid_reference)"); Hope this may help someone in the future facing the same issue. Regards, Grant Bagdasarian CM

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

sr-users